BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//IDSS STAGE - ECPv6.15.11//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:IDSS STAGE
X-ORIGINAL-URL:https://idss-stage.mit.edu
X-WR-CALDESC:Events for IDSS STAGE
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20160313T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20161106T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20170312T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20171105T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20180311T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20181104T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20190310T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20191103T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20200308T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20201101T060000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20190409T160000
DTEND;TZID=America/New_York:20190409T170000
DTSTAMP:20260524T190854
CREATED:20190301T171340Z
LAST-MODIFIED:20190501T142316Z
UID:8985-1554825600-1554829200@idss-stage.mit.edu
SUMMARY:Personalized Dynamic Pricing with Machine Learning: High Dimensional Covariates and Heterogeneous Elasticity
DESCRIPTION:We consider a seller who can dynamically adjust the price of a product at the individual customer level\, by utilizing information about customers’ characteristics encoded as a $d$-dimensional feature vector. We assume a personalized demand model\, parameters of which depend on $s$ out of the $d$ features. The seller initially does not know the relationship between the customer features and the product demand\, but learns this through sales observations over a selling horizon of $T$ periods. We prove that the seller’s expected regret\, i.e.\, the revenue loss against a clairvoyant who knows the underlying demand relationship\, is at least of order $s\sqrt{T}$ under any admissible policy. We then design a near-optimal pricing policy for a “semi-clairvoyant” seller (who knows which s of the d features are in the demand model) that achieves an expected regret of order $s\sqrt{T}log(T)$. We extend this policy to a more realistic setting where the seller does not know the true demand predictors\, and show this policy has an expected regret of order $s\sqrt{T}(log(d)＋log(T))$\, which is also near-optimal. Finally\, we test our theory on simulated data and on a data set from an online auto loan company in the United States. On both data sets\, our experimentation-based pricing policy is superior to intuitive and/or widely-practiced customized pricing methods such as myopic pricing and segment-then-optimize policies. Furthermore\, our policy significantly improves upon the loan company’s historical pricing decisions in terms of annual expected revenue. \nBio: Gah-Yi Ban is an Assistant Professor of Management Science and Operations at London Business School. Gah-Yi’s research is in Big Data analytics\, specifically decision-making with complex\, high-dimensional and/or highly uncertain data with applications to operations management and finance. Gah-Yi’s research has appeared on most-downloaded lists of Management Science and Operations Research\, and awarded Honorable Mention in 2018 INFORMS JFIG Paper Competition. Gah-Yi graduated from UC Berkeley with MSc/MA/PhD in Industrial Engineering/ Statistics/Operations Research. \n____________________________________ \nThe LIDS Seminar Series features distinguished speakers who provide an overview of a research area\, as well as exciting recent progress in that area. Intended for a broad audience\, seminar topics span the areas of communications\, computation\, control\, learning\, networks\, probability and statistics\, optimization\, and signal processing. 
URL:https://idss-stage.mit.edu/calendar/lids-seminar-gah-yi-ban-london-business-school/
LOCATION:32-155
CATEGORIES:LIDS Seminar Series
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20190312T160000
DTEND;TZID=America/New_York:20190312T170000
DTSTAMP:20260524T190854
CREATED:20190301T170833Z
LAST-MODIFIED:20190501T142423Z
UID:8981-1552406400-1552410000@idss-stage.mit.edu
SUMMARY:Automatic Computation of Exact Worst-Case Performance for First-Order Methods
DESCRIPTION:Joint work with Adrien Taylor (INRIA) and Francois Glineur (UCLouvain). \nWe show that the exact worst-case performances of a wide class of first-order convex optimization algorithms can be obtained as solutions to semi-definite programs\, which provide both the performance bounds and functions on which these are reached.  Our formulation is based on a necessary and sufficient condition for smooth (strongly) convex interpolation\, allowing for a finite representation for smooth (strongly) convex functions in this context. These results allow improving the performance bounds of many classical algorithms\, and better understanding their dependence on the algorithm’s parameters\, leading to new optimized parameters\, and thus stronger performances. \nOur approach can be applied via the PESTO Toolbox\, which let the user describe algorithms in a natural way. \nBio: Julien M. Hendrickx is professor of mathematical engineering at Université catholique de Louvain\, in the Ecole Polytechnique de Louvain since 2010. He is on sabbatical at Boston University in 2018-19\, holding a WBI-World excellence fellowship. \nHe obtained an engineering degree in applied mathematics (2004) and a PhD in mathematical engineering (2008) from the same university. He has been a visiting researcher at the University of Illinois at Urbana Champaign in 2003-2004\, at the National ICT Australia in 2005 and 2006\, and at the Massachusetts Institute of Technology in 2006 and 2008. He was a postdoctoral fellow at the Laboratory for Information and Decision Systems of the Massachusetts Institute of Technology 2009 and 2010\, holding postdoctoral fellowships of the F.R.S.-FNRS (Fund for Scientific Research) and of Belgian American Education Foundation. \nDoctor Hendrickx is the recipient of the 2008 EECI award for the best PhD thesis in Europe in the field of Embedded and Networked Control\, and of the Alcatel-Lucent-Bell 2009 award for a PhD thesis on original new concepts or application in the domain of information or communication technologies. \n____________________________________ \nThe LIDS Seminar Series features distinguished speakers who provide an overview of a research area\, as well as exciting recent progress in that area. Intended for a broad audience\, seminar topics span the areas of communications\, computation\, control\, learning\, networks\, probability and statistics\, optimization\, and signal processing. 
URL:https://idss-stage.mit.edu/calendar/automatic-computation-of-exact-worst-case-performance-for-first-order-methods/
LOCATION:32-155
CATEGORIES:LIDS Seminar Series
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20190226T160000
DTEND;TZID=America/New_York:20190226T170000
DTSTAMP:20260524T190854
CREATED:20190301T170447Z
LAST-MODIFIED:20190501T142448Z
UID:8979-1551196800-1551200400@idss-stage.mit.edu
SUMMARY:Coded Computing: A Transformative Framework for Resilient\, Secure\, and Private Distributed Learning
DESCRIPTION:This talk introduces “Coded Computing”\, a new framework that brings concepts and tools from information theory and coding into distributed computing to mitigate several performance bottlenecks that arise in large-scale distributed computing and machine learning\, such as resiliency to stragglers and bandwidth bottleneck. Furthermore\, coded computing can enable (information-theoretically) secure and private learning over untrusted workers that is gaining increasing importance in various application domains. In particular\, we present CodedPrivateML for distributed learning\, which keeps both the data and the model private while allowing efficient parallelization of training across untrusted distributed workers. We demonstrate that CodedPrivateML can provide an order of magnitude speedup (up to ~30x) over the cryptographic approaches that rely on secure multiparty computing. \nBio: Salman Avestimehr is a Professor of Electrical Engineering and co-director of Communication Sciences Institute at the University of Southern California. He received his Ph.D. in 2008 and M.S. degree in 2005 in Electrical Engineering and Computer Science\, both from the University of California\, Berkeley. Prior to that\, he obtained his B.S. in Electrical Engineering from Sharif University of Technology in 2003.  His research interests include information theory and coding\, distributed computing\, and machine learning. Dr. Avestimehr has received a number of awards\, including a Communications Society and Information Theory Society Joint Paper Award\, the Presidential Early Career Award for Scientists and Engineers (PECASE)\, a Young Investigator Program (YIP) award from the U. S. Air Force Office of Scientific Research\, a National Science Foundation CAREER award\, and several best paper awards. He is currently an Associate Editor for the IEEE Transactions on Information Theory and a General Co-Chair of the 2020 International Symposium on Information Theory (ISIT). \n____________________________________ \nThe LIDS Seminar Series features distinguished speakers who provide an overview of a research area\, as well as exciting recent progress in that area. Intended for a broad audience\, seminar topics span the areas of communications\, computation\, control\, learning\, networks\, probability and statistics\, optimization\, and signal processing. 
URL:https://idss-stage.mit.edu/calendar/coded-computing-a-transformative-framework-for-resilient-secure-and-private-distributed-learning/
LOCATION:32-155
CATEGORIES:LIDS Seminar Series
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20190219T160000
DTEND;TZID=America/New_York:20190219T170000
DTSTAMP:20260524T190854
CREATED:20190301T165622Z
LAST-MODIFIED:20190501T142520Z
UID:8976-1550592000-1550595600@idss-stage.mit.edu
SUMMARY:Safeguarding Privacy in Dynamic Decision-Making Problems
DESCRIPTION:The increasing ubiquity of large-scale infrastructures for surveillance and data analysis has made understanding the impact of privacy a pressing priority in many domains. We propose a framework for studying a fundamental cost vs. privacy tradeoff in dynamic decision-making problems. The central question is: how can a decision maker take actions that are efficient for her goal\, while simultaneously ensuring these actions do not inadvertently reveal her private information\, even when observed and analyzed by a powerful adversary? We will examine two well-known decision problems (path planning and online learning)\, and in both cases establish sharp\, information-theoretic complexity vs. privacy tradeoff. As a by-product\, our analysis also leads to simple yet provably efficient algorithms for both the decision maker and eavesdropping adversary. Based in part on joint work with John N. Tsitsiklis and Zhi Xu (MIT). \nBio: Kuang Xu was born in Suzhou\, China. He is an Assistant Professor of Operations\, Information and Technology at the Stanford Graduate School of Business\, Stanford University. He received the B.S. degree in Electrical Engineering (2009) from the University of Illinois at Urbana-Champaign\, Urbana\, Illinois\, USA\, and the Ph.D. degree in Electrical Engineering and Computer Science (2014) from the Massachusetts Institute of Technology\, Cambridge\, Massachusetts\, USA. His research interests lie in the fields of applied probability theory\, optimization\, and operations research\, seeking to understand fundamental properties and design principles of large-scale stochastic systems\, with applications in queueing networks\, healthcare\, privacy\, and statistical learning theory. He has received several awards including a First Place in INFORMS George E. Nicholson Student Paper Competition\, a Best Paper Award\, as well as a Kenneth C. Sevcik Outstanding Student Paper Award from ACM SIGMETRICS. \n____________________________________ \nThe LIDS Seminar Series features distinguished speakers who provide an overview of a research area\, as well as exciting recent progress in that area. Intended for a broad audience\, seminar topics span the areas of communications\, computation\, control\, learning\, networks\, probability and statistics\, optimization\, and signal processing. 
URL:https://idss-stage.mit.edu/calendar/kuang-xu/
LOCATION:32-155
CATEGORIES:LIDS Seminar Series
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20181210T160000
DTEND;TZID=America/New_York:20181210T170000
DTSTAMP:20260524T190854
CREATED:20180810T161120Z
LAST-MODIFIED:20190501T143433Z
UID:8174-1544457600-1544461200@idss-stage.mit.edu
SUMMARY:Symmetry\, Bifurcation\, and Multi-Agent Decision-Making
DESCRIPTION:Prof. Leonard will present nonlinear dynamics for distributed decision-making that derive from principles of symmetry and bifurcation. Inspired by studies of animal groups\, including house-hunting honeybees and schooling fish\, the nonlinear dynamics describe a group of interacting agents that can manage flexibility as well as stability in response to a changing environment. \nBio: Prof. Naomi Ehrich Leonard is Edwin S. Wilsey Professor of Mechanical and Aerospace Engineering and associated faculty in Applied and Computational Mathematics at Princeton University. She is a MacArthur Fellow\, and Fellow of the American Academy of Arts and Sciences\, SIAM\, IEEE\, IFAC\, and ASME. She received her BSE in Mechanical Engineering from Princeton University and her PhD in Electrical Engineering from the University of Maryland. Her research is in control and dynamics with application to multi-agent systems\, mobile sensor networks\, collective animal behavior\, and human decision dynamics. \n____________________________________ \nThe LIDS Seminar Series features distinguished speakers who provide an overview of a research area\, as well as exciting recent progress in that area. Intended for a broad audience\, seminar topics span the areas of communications\, computation\, control\, learning\, networks\, probability and statistics\, optimization\, and signal processing. 
URL:https://lids.mit.edu/news-and-events/events/lids-seminar-series-6
LOCATION:32-155
CATEGORIES:LIDS Seminar Series
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20181126T160000
DTEND;TZID=America/New_York:20181126T170000
DTSTAMP:20260524T190854
CREATED:20180810T160857Z
LAST-MODIFIED:20190501T143456Z
UID:8172-1543248000-1543251600@idss-stage.mit.edu
SUMMARY:Transportation Systems Resilience: Capacity-Aware Control and Value of Information
DESCRIPTION:Resilience of a transportation system is its ability to operate under adverse events like incidents and storms. Availability of real-time traffic data provides new opportunities for predicting travelers’ routing behavior and implementing network control operations during adverse events. In this talk\, we will discuss two problems: controlling highway corridors in response to disruptions and modeling strategic route choices of travelers with heterogeneous access to incident information. Firstly\, we present an approach to designing control strategies for highway corridors facing stochastic capacity disruptions such random incidents and vehicle platoons/moving bottlenecks. We exploit the properties of traffic flow dynamics under recurrent incidents to derive verifiable conditions for stability of traffic queues\, and also obtain guarantees on the system throughput. Secondly\, we introduce a routing game in which travelers receive asymmetric and incomplete information about uncertain network state\, and make route choices based on their private beliefs about the state and other travelers’ behavior. We study the effects of information heterogeneity on travelers’ equilibrium route choices and costs. Our analysis is useful for evaluating the value of receiving state information for travelers\, which can be positive\, zero\, or negative in equilibrium. These results demonstrate the advantages of considering network state uncertainty in both strategic and operational aspects of system resilience. \nBio: Saurabh Amin is Robert N. Noyce Career Development Associate Professor in the Department of Civil and Environmental Engineering at MIT. He is also affiliated with the Institute of Data\, Systems and Society and the Operations Research Center at MIT. His research focuses on the design of network inspection and control algorithms for infrastructure systems resilience. He studies the effects of security attacks and natural events on the survivability of cyber-physical systems\, and designs incentive mechanisms to reduce network risks. Dr. Amin received his Ph.D. from the University of California\, Berkeley in 2011. His research is supported by NSF CPS FORCES Frontiers project\, NSF CAREER award\, Google Faculty Research award\, DoD-Science of Security Program\, and Siebel Energy Institute Grant. \n____________________________________ \nThe LIDS Seminar Series features distinguished speakers who provide an overview of a research area\, as well as exciting recent progress in that area. Intended for a broad audience\, seminar topics span the areas of communications\, computation\, control\, learning\, networks\, probability and statistics\, optimization\, and signal processing. 
URL:https://lids.mit.edu/news-and-events/events/lids-seminar-series-saurabh-amin
LOCATION:32-155
CATEGORIES:LIDS Seminar Series
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20181119T160000
DTEND;TZID=America/New_York:20181119T170000
DTSTAMP:20260524T190854
CREATED:20180810T160618Z
LAST-MODIFIED:20190501T143517Z
UID:8170-1542643200-1542646800@idss-stage.mit.edu
SUMMARY:Modeling Electricity Markets with Complementarity: Why It's Important (and Fun)
DESCRIPTION:Electric power: done wrong\, it drags the economy and environment down; done right\, it could help to create a more efficient\, brighter\, and cleaner future. Better policy\, planning\, and operations models–both simple analytical\, and complex computational ones–are essential if we’re going to do it right. Better modeling is also fun\, as the math of electricity models is inherently interesting and revealing –models often show flaws in our intuition. Used intelligently\, models can point us towards better regulations\, investments\, and operating policies. Simple models provide insights\, while complex models provide the numbers needed to choose specific investments and policies. \n\nComplementarity is one optimization-based approach to modeling energy markets that has more flexibility to model market failures than standard optimization methods. Prof. Hobbs will highlight one application using the power market model COMPETES: the design of renewable portfolio standards\, and an analysis of their price and economic efficiency impacts in the Year 2030. The focus is on energy versus capacity subsidies in the European Union; capacity subsidies are being promoted as potentially being more effective in promoting technology learning. They also have less of an impact upon electricity prices. Prof. Hobbs will also examine the cost of country-specific targets versus EU-wide targets. \nAcknowledgments: Government of the Netherlands and NSF for funding; my PBL colleagues Ozge Ozdemir\, Paul Koustaal\, and Marit van Hout. \n\nBio: B.F. Hobbs earned a Ph.D. (Environmental Systems Engineering) in 1983 from Cornell University. He holds the Theodore M. and Kay W. Schad Chair of Environmental Management at the Johns Hopkins University\, where he has been in the Department of Geography & Environmental Engineering (now Environmental Health & Engineering) since 1995. He also holds a joint appointment in the Department of Applied Mathematics & Statistics and is founding director of the JHU Environment\, Energy\, Sustainability & Health Institute. He co-directs the EPA Yale-JHU Center for Solutions for Energy\, Air\, Climate and Health (SEArCH). Previously\, he was at Brookhaven and Oak Ridge National Laboratories and a member of the Systems Engineering and Civil Engineering faculty at Case Western Reserve University. \nHis research and teaching concern the application of systems analysis and economics to electric utility regulation\, planning\, and operations\, as well as environmental and water resources systems. Dr. Hobbs has previously held visiting appointments at CalTech\, Comillas Pontifical University\, Helsinki University of Technology\, University of Washington\, Netherlands Energy Research Center\, and Cambridge University. He chairs the Market Surveillance Committee of the California Independent System Operator. He was named an NSF Presidential Young Investigator in 1986. Dr. Hobbs is a Fellow of the IEEE and INFORMS. \n____________________________________ \nThe LIDS Seminar Series features distinguished speakers who provide an overview of a research area\, as well as exciting recent progress in that area. Intended for a broad audience\, seminar topics span the areas of communications\, computation\, control\, learning\, networks\, probability and statistics\, optimization\, and signal processing. 
URL:https://lids.mit.edu/news-and-events/events/lids-seminar-series-5
LOCATION:32-155
CATEGORIES:LIDS Seminar Series
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20181113T080000
DTEND;TZID=America/New_York:20181113T170000
DTSTAMP:20260524T190854
CREATED:20180810T160435Z
LAST-MODIFIED:20190501T143540Z
UID:8168-1542096000-1542128400@idss-stage.mit.edu
SUMMARY:Functional Representation of Random variables and Applications
DESCRIPTION:The functional representation lemma says that given random variables X and Y\, there exists a random variable Z\, independent of X\, and a function g(x\,z) such that Y=g(X\,Z). This lemma has had several applications in information theory aimed at simplifying computations of certain information functional. I will present a strengthened version of this lemma and applications to several one-shot coding problems. The first application is to channel simulation with common randomness\, where we obtain an improved bound on the achievable rate by Harsha et al. that applies to arbitrary (not just discrete) random variables. More interestingly\, the Poisson construction used in the proof of the strengthened lemma leads to new and simple achievability results for one-shot coding theorems\, including lossy source coding\, multiple description coding\, and the Gray–Wyner system. I will end with an application of the Poisson construction to minimax learning for remote inference. \nThe new results presented in this talk are joint with Cheuk Ting Li\, Xiugang Wu\, and Ayfer Ozgur. \nBio: Abbas El Gamal is the Hitachi America Professor in the School of Engineering at Stanford University. He received his B.Sc. Honors degree from Cairo University in 1972\, and his M.S. in Statistics and Ph.D. in Electrical Engineering both from Stanford University in 1977 and 1978\, respectively. From 1978 to 1980\, he was an Assistant Professor of Electrical Engineering at USC. From 2003 to 2012\, he was Director of the Information Systems Laboratory at Stanford University. From 2012-2017 he was the Fortinet Founders Chair of the Department of Electrical Engineering at Stanford University. His research contributions have been in network information theory\, FPGAs\, digital imaging devices and systems\, and smart grid modeling and control. He has authored or coauthored over 230 papers and holds 35 patents in these areas. He is a coauthor of the book Network Information Theory (Cambridge Press 2011). He is a member of the US National Academy of Engineering and a Fellow of the IEEE. He received several awards for his research contributions\, including the 2016 IEEE Richard Hamming Medal and the 2012 Claude E. Shannon Award. He served on the Board of Governors of the Information Theory Society from 2009 to 2016 and was its President in 2014. He has been involved in several Silicon Valley startups as co-founder\, a board of director member\, advisor and in several key technical and management positions. \n\n____________________________________ \nThe LIDS Seminar Series features distinguished speakers who provide an overview of a research area\, as well as exciting recent progress in that area. Intended for a broad audience\, seminar topics span the areas of communications\, computation\, control\, learning\, networks\, probability and statistics\, optimization\, and signal processing. 
URL:https://lids.mit.edu/news-and-events/events/lids-seminar-series-abbas-el-gamal
LOCATION:32-141\, United States
CATEGORIES:LIDS Seminar Series
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20181029T160000
DTEND;TZID=America/New_York:20181029T170000
DTSTAMP:20260524T190854
CREATED:20180810T160254Z
LAST-MODIFIED:20190501T143409Z
UID:8166-1540828800-1540832400@idss-stage.mit.edu
SUMMARY:Computing with Assemblies
DESCRIPTION:Computation in the brain has been modeled productively at many scales\, ranging from molecules to dendrites\, neurons\, and synapses\, all the way to the whole brain models useful in cognitive science. I will discuss recent work on an intermediate layer\, involving assemblies of neurons — that is to say\, sets of neurons firing together in a repetitive pattern whenever we think of a particular memory\, concept or idea. Assemblies have been conjectured six decades ago by Hebb\, and have been over the past decade noticed in both the animal and the human brain. Further\, experiments\, simulations\, and theoretical analysis suggest that assemblies can be copied from one brain area to another\, and associated with other assemblies to encode affinity. We propose a broader “calculus” of assemblies\, including operations such as “reciprocal-project” and “merge”\, comprising a powerful computational model. One interesting hypothesis is that assembly operations may underlie some of the most advanced functions of the brain\, such as reasoning\, planning\, language\, math. Work with Santosh Vempala\, Wolfgang Maass\, and Michael Collins. \n\n\nBio: Christos H. Papadimitriou is the Donovan Family professor of computer science at Columbia University. Before joining Columbia in 2017\, he taught at UC Berkeley for 22 years\, and before that at Harvard\, MIT\, NTU Athens\, Stanford\, and UCSD. He has written five textbooks and many articles on algorithms and complexity\, and their applications to optimization\, databases\, control\, AI\, robotics\, economics and game theory\, the Internet\, evolution\, and more recently the study of the brain. He holds a Ph.D. from Princeton as well as eight honorary doctorates\, and he has won the Knuth prize\, the Goedel prize\, and the von Neumann Medal. He is a member of the National Academy of Sciences of the US\, the American Academy of Arts and Sciences\, and the National Academy of Engineering; in 2013 the president of Greece named him Commander of the Order of the Phoenix. He has also written three novels: “Turing”\, “Logicomix” and his latest “Independence”. \n\n____________________________________ \nThe LIDS Seminar Series features distinguished speakers who provide an overview of a research area\, as well as exciting recent progress in that area. Intended for a broad audience\, seminar topics span the areas of communications\, computation\, control\, learning\, networks\, probability and statistics\, optimization\, and signal processing. 
URL:https://lids.mit.edu/news-and-events/events/lids-seminar-series-4
LOCATION:32-155
CATEGORIES:LIDS Seminar Series
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20181022T160000
DTEND;TZID=America/New_York:20181022T170000
DTSTAMP:20260524T190854
CREATED:20180810T160117Z
LAST-MODIFIED:20190501T143713Z
UID:8164-1540224000-1540227600@idss-stage.mit.edu
SUMMARY:Distributed Statistical Estimation of High-Dimensional Distributions and Parameters under Communication Constraints
DESCRIPTION:Modern data sets are often distributed across multiple machines and processors\, and bandwidth and energy limitations in networks and within multiprocessor systems often impose significant bottlenecks on the performance of algorithms. Motivated by this trend\, we consider the problem of estimating high-dimensional distributions and parameters in a distributed network\, where each node in the network observes an independent sample from the underlying distribution and can communicate it to a central processor by writing at most k bits on a public blackboard. We obtain matching upper and lower bounds for the minimax risk of estimating the underlying distribution or parameter under various common statistical models. Our results show that the impact of the communication constraint can be qualitatively different depending on the tail behavior of the score function associated with each model. The key ingredient in our proof is a geometric characterization of Fisher information from quantized samples. \nJoint work with Leighton Barnes\, Yanjun Han\, and Tsachy Weissman. \nBio: Ayfer Ozgur received her Ph.D. degree in 2009 from the Information Processing Group at EPFL\, Switzerland. In 2010 and 2011\, she was a post-doctoral scholar at the same institution. She is an Assistant Professor in the Electrical Engineering Department at Stanford University since 2012. Her research interests include distributed communication and learning\, wireless systems\, and information theory. Dr. Ozgur received the EPFL Best Ph.D. Thesis Award in 2010\, an NSF CAREER award in 2013\, the Okawa Foundation Research Grant and the IEEE Communication Theory Technical Committee (CTTC) Early Achievement Award in 2018. \n\n____________________________________ \nThe LIDS Seminar Series features distinguished speakers who provide an overview of a research area\, as well as exciting recent progress in that area. Intended for a broad audience\, seminar topics span the areas of communications\, computation\, control\, learning\, networks\, probability and statistics\, optimization\, and signal processing. 
URL:https://lids.mit.edu/news-and-events/events/lids-seminar-series-3
LOCATION:32-155
CATEGORIES:LIDS Seminar Series
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20181015T160000
DTEND;TZID=America/New_York:20181015T170000
DTSTAMP:20260524T190854
CREATED:20180810T155930Z
LAST-MODIFIED:20190501T143758Z
UID:8162-1539619200-1539622800@idss-stage.mit.edu
SUMMARY:Augmented Lagrangians and Decomposition in Convex and Nonconvex Programming
DESCRIPTION:Multiplier methods based on augmented Lagrangians are attractive in convex and nonconvex programming for their stabilizing and even convexifying properties. They have widely been seen\, however\, as incompatible with taking advantage of a block-separable structure. \nIn fact\, when articulated in the right way\, they can produce decomposition algorithms in which low-dimensional subproblems can be solved in parallel. Convergence in the nonconvex case is\, of course\, just local\, but is available under a broad analog of the strong second-order sufficient condition for local optimality that dominates much of computational methodology outside of convex optimization. This carries over also to extended nonlinear programming with its greater flexibility to handle composite terms. \nBio: Ralph Tyrrell (Terry) Rockafellar has long been associated with the University of Washington\, Seattle\, where he is Professor Emeritus of Mathematics\, but has also contributed in recent years as Adjunct Research Professor of Systems and Industrial Engineering at the University of Florida\, Gainesville\, and as Honorary Professor of the Department of Applied Mathematics at Hong Kong Polytechnic University. \nHis interests span from convex and variational analysis to problems of optimization and equilibrium\, especially nowadays applications in finance\, engineering\, and economics involving risk and reliability\, along with schemes of problem decomposition on convex and nonconvex programming. \nIn addition to being a winner of the Dantzig Prize given jointly by SIAM and the Mathematical Programming Society (1983)\, Prof. Rockafellar has gained international recognition for his work through honorary doctorates bestowed by universities in a number of countries. INFORMS awarded him and Roger Wets the 1997 Lancaster Prize for their book Variational Analysis\, and in 1999 he was further honored by INFORMS with John von Neumann Theory Prize for his fundamental contributions to the methodology of optimization. He has authored over 240 publications\, including one of the all-time most highly cited books in mathematics\, Convex Analysis. \n____________________________________ \nThe LIDS Seminar Series features distinguished speakers who provide an overview of a research area\, as well as exciting recent progress in that area. Intended for a broad audience\, seminar topics span the areas of communications\, computation\, control\, learning\, networks\, probability and statistics\, optimization\, and signal processing. 
URL:https://lids.mit.edu/news-and-events/events/lids-seminar-series-2
LOCATION:32-155
CATEGORIES:LIDS Seminar Series
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20180924T160000
DTEND;TZID=America/New_York:20180924T170000
DTSTAMP:20260524T190854
CREATED:20180810T155456Z
LAST-MODIFIED:20190501T143856Z
UID:8160-1537804800-1537808400@idss-stage.mit.edu
SUMMARY:The Power of Multiple Samples in Generative Adversarial Networks
DESCRIPTION:We bring the tools from Blackwell’s seminal result on comparing two stochastic experiments from 1953\, to shine a new light on a modern application of great interest: Generative Adversarial Networks (GAN). Binary hypothesis testing is at the center of training GANs\, where a trained neural network (called a critic) determines whether a given sample is from the real data or the generated (fake) data. By jointly training the generator and the critic\, the hope is that eventually the trained generator will generate realistic samples. One of the major challenges in GAN is known as “mode collapse”; the lack of diversity in the samples generated by thus trained generators. We propose a new training framework\, where the critic is fed with multiple samples jointly (which we call packing)\, as opposed to each sample separately as done in standard GAN training. With this simple but fundamental departure from standard GANs\, experimental results show that the diversity of the generated samples improve significantly. We analyze this practical gain by first providing a formal mathematical definition of mode collapse and making a fundamental connection between the idea of packing and the intensity of mode collapse. Precisely\, we show that the packed critic naturally penalizes mode collapse\, thus encouraging generators with less mode collapse. The analyses critically rely on operational interpretation of hypothesis testing and corresponding data processing inequalities\, which lead to sharp analyses with simple proofs. For this talk\, I will assume no prior background on GANs. \nThis is joint work with Zinan Lin (CMU)\, Ahsish Khetan (Amazon AI)\, and Giulia Fanti (CMU). \nBio: Sewoong Oh is an Associate Professor of Industrial and Enterprise Systems Engineering at UIUC. He received his PhD from the department of Electrical Engineering at Stanford University. Following his PhD\, he worked as a postdoctoral researcher at Laboratory for Information and Decision Systems (LIDS) at MIT. His research interest is in theoretical machine learning\, including spectral methods\, ranking\, crowdsourcing\, estimation of information measures\, differential privacy\, and generative adversarial networks. He was co-awarded the best paper award at the SIGMETRICS in 2015\, NSF CAREER award in 2016 and GOOGLE Faculty Research Award. \n____________________________________ \nThe LIDS Seminar Series features distinguished speakers who provide an overview of a research area\, as well as exciting recent progress in that area. Intended for a broad audience\, seminar topics span the areas of communications\, computation\, control\, learning\, networks\, probability and statistics\, optimization\, and signal processing. 
URL:https://lids.mit.edu/news-and-events/events/lids-seminar-series-1
LOCATION:32-155
CATEGORIES:LIDS Seminar Series
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20180917T160000
DTEND;TZID=America/New_York:20180917T170000
DTSTAMP:20260524T190854
CREATED:20180810T155258Z
LAST-MODIFIED:20190501T143950Z
UID:8158-1537200000-1537203600@idss-stage.mit.edu
SUMMARY:Regret of Queueing Bandits
DESCRIPTION:We consider a variant of the multiarmed bandit (MAB) problem where jobs or tasks queue for service\, and service rates of different servers (agents) may be unknown. Such (queueing+learning) problems are motivated by a vast range of service systems\, including supply and demand in online platforms (e.g.\, Uber\, Lyft\, Airbnb\, Upwork\, etc.)\, order flow in financial markets (e.g.\, limit order books)\, communication systems\, and supply chains. \nWe study algorithms that minimize queue-regret: the expected difference between the queue-lengths (backlogs) obtained by the algorithm\, and those obtained by a genie-aided matching algorithm that knows exact service rates. A naive view of this problem would suggest that queue-regret could grow logarithmically: since queue-regret cannot be larger than classical regret\, results for the standard MAB problem give algorithms that ensure queue-regret increases no more than logarithmically in time. Our work shows surprisingly more complex behavior — specifically\, the optimal queue-regret decreases with time and scales as O(1/t). We next consider holding-cost regret in multi-class (multiple types of tasks) multi-server (servers/agents have task-type dependent service rate) systems. Holding costs correspond to a system where a linear cost (with respect to time spent in the queue) is incurred for each incomplete task. We consider learning-based variants of the c-mu rule – a classic and well-studied scheduling policy that is used when server/agent service rates are known. We develop algorithms that result in constant expected holding-cost regret (independent of time). The key insight that allows such a regret bound is that service systems we consider exhibit explore-free learning\, where no penalty is (eventually) incurred for exploring and learning server/agent rates. We finally discuss the implications of our results on building platforms for matching tasks to servers/agents. Base on joint work with Subhashini Krishnasamy\, Rajat Sen\, Ari Arapostathis and Ramesh Johari. \nBio: Sanjay Shakkottai received his Ph.D. from the ECE Department at the University of Illinois at Urbana-Champaign in 2002. He is with The University of Texas at Austin\, where he is currently the Ashley H. Priddy Centennial Professor in Engineering\, the Director of the Wireless Networking and Communications Group (WNCG)\, and a Professor in the Department of Electrical and Computer Engineering. He received the NSF CAREER award in 2004 and was elected as an IEEE Fellow in 2014. His research interests lie at the intersection of algorithms for resource allocation\, statistical learning and networks\, with applications to wireless communication networks and online platforms. \n____________________________________ \nThe LIDS Seminar Series features distinguished speakers who provide an overview of a research area\, as well as exciting recent progress in that area. Intended for a broad audience\, seminar topics span the areas of communications\, computation\, control\, learning\, networks\, probability and statistics\, optimization\, and signal processing. 
URL:https://lids.mit.edu/news-and-events/events/lids-seminar-series-0
LOCATION:32-155
CATEGORIES:LIDS Seminar Series
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20180910T160000
DTEND;TZID=America/New_York:20180910T170000
DTSTAMP:20260524T190854
CREATED:20180905T161016Z
LAST-MODIFIED:20190501T144029Z
UID:8230-1536595200-1536598800@idss-stage.mit.edu
SUMMARY:Streaming Analytics for the Smart Grid
DESCRIPTION:How to conduct real-time analytics of streaming measurement data in the power grid? This talk offers a dynamic systems approach to utilizing data of different time scale for improved monitoring of the grid cyber and physical security. The first example of the talk presents how to leverage synchrophasor data dimensionality reduction and Robust Principal Component Analysis for early anomaly detection\, visualization\, and localization. The second example presents an online framework to detect cyber-attacks on automatic generation control (AGC). A cyber-attack detection algorithm is designed based on the approach of Dynamic Watermarking. The detection algorithm provides a theoretical guarantee of detection of cyber-attacks launched by sophisticated attackers possessing extensive knowledge of the physical and statistical models of targeted power systems. The underlying theme of the work suggests the importance of integrating data with dynamic context-aware models in the smart grid. \nBio: Dr. Le Xie is a Professor and Eugene Webb Faculty Fellow in the Department of Electrical and Computer Engineering at Texas A&M University. He received B.E. in Electrical Engineering from Tsinghua University\, S.M. in Engineering Sciences from Harvard\, and Ph.D. in Electrical and Computer Engineering from Carnegie Mellon in 2009. His industry experience includes ISO-New England and Edison Mission Energy Marketing and Trading. His research interest includes modeling and control in data-rich large-scale systems\, grid integration of clean energy resources\, and electricity markets. \nDr. Xie received the U.S. National Science Foundation CAREER Award\, and DOE Oak Ridge Ralph E. Powe Junior Faculty Enhancement Award. He was awarded the 2017 IEEE PES Outstanding Young Engineer Award. He was the recipient of Texas A&M Dean of Engineering Excellence Award\, ECE Outstanding Professor Award\, and TEES Select Young Fellow. He is an Editor of IEEE Transactions on Smart Grid\, and the founding chair of IEEE PES Subcommittee on Big Data & Analytics for Grid Operations. He and his students received the Best Paper awards at North American Power Symposium and IEEE SmartGridComm. He recently chaired the 2018 NSF Workshop on Real-time Learning and Decision Making in Dynamical Systems. \n____________________________________ \nThe LIDS Seminar Series features distinguished speakers who provide an overview of a research area\, as well as exciting recent progress in that area. Intended for a broad audience\, seminar topics span the areas of communications\, computation\, control\, learning\, networks\, probability and statistics\, optimization\, and signal processing. 
URL:https://lids.mit.edu/news-and-events/events/streaming-analytics-smart-grid
LOCATION:32-155
CATEGORIES:LIDS Seminar Series
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20180515T150000
DTEND;TZID=America/New_York:20180515T160000
DTSTAMP:20260524T190854
CREATED:20180223T173133Z
LAST-MODIFIED:20180515T135702Z
UID:7445-1526396400-1526400000@idss-stage.mit.edu
SUMMARY:LIDS Seminar Series - Vivek Borkar
DESCRIPTION:
URL:https://idss-stage.mit.edu/calendar/https-lids-mit-edu-news-and-events-events-distributed-algorithms-tsitsiklis-and-beyond/
LOCATION:32-141\, United States
CATEGORIES:LIDS Seminar Series
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20180508T150000
DTEND;TZID=America/New_York:20180508T160000
DTSTAMP:20260524T190854
CREATED:20180223T173042Z
LAST-MODIFIED:20180223T173042Z
UID:7443-1525791600-1525795200@idss-stage.mit.edu
SUMMARY:A Rationally Designed Biomolecular Integral Feedback Control System for Robust Gene Regulation
DESCRIPTION:Abstract \nHumans have been influencing the DNA of plants and animals for thousands of years through selective breeding. Yet it is only over the last 3 decades or so that we have gained the ability to manipulate the DNA itself and directly alter its sequences through the modern tools of genetic engineering. This has revolutionized biotechnology and ushered in the era of synthetic biology. Among the possible applications enabled by synthetic biology is the design and engineering of feedback control systems that act at the molecular scale in real-time to steer the dynamic behavior of living cells. Here I will present our theoretical framework for the design and synthesis of such control systems\, and will discuss the main challenges in their practical implementation. I will then present the first designer gene network that attains integral feedback in a living cell and demonstrate its tenability and disturbance rejection properties. A growth control application shows the inherent capacity of this integral feedback control system to deliver robustness\, and highlights its potential use as a universal controller for regulation of biological variables in arbitrary networks. Finally\, I will discuss the potential impact of biomolecular control systems in industrial biotechnology and medical therapy and bring attention to the opportunities that exist for control theorists to advance this young area of research. \nBiography \nMustafa Khammash is the Professor of Control Theory and Systems Biology at the Department of Biosystems Science and Engineering at ETH Zurich\, Switzerland. He works in the areas of control theory\, systems biology\, and synthetic biology. His lab develops theoretical\, computational\, and experimental methods aimed at understanding the role of dynamics\, feedback\, and randomness in biology. He is currently developing new theoretical and experimental approaches for the design of biomolecular control systems and for their realization in living cells. \nProf. Khammash received his B.S. degree from Texas A&M University in 1986 and his Ph.D. from Rice University in 1990\, both in electrical engineering. In 1990\, he joined the engineering faculty of Iowa State University\, where he created the Dynamics and Control Program and led the control group until 2002. He then joined the engineering faculty at the University of California\, Santa Barbara (UCSB)\, where he was Director of the Center for Control\, Dynamical Systems and Computation (CCDC) until 2011 when he joined ETH Zurich. He is a Fellow of the IEEE\, IFAC\, and the Japan Society for the Promotion of Science (JSPS).
URL:https://idss-stage.mit.edu/calendar/a-rationally-designed-biomolecular-integral-feedback-control-system-for-robust-gene-regulation/
LOCATION:32-141\, United States
CATEGORIES:LIDS Seminar Series
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20180424T150000
DTEND;TZID=America/New_York:20180424T160000
DTSTAMP:20260524T190854
CREATED:20180223T172912Z
LAST-MODIFIED:20180419T194543Z
UID:7441-1524582000-1524585600@idss-stage.mit.edu
SUMMARY:LIDS Seminar Series: Jose M. F. Moura
DESCRIPTION:
URL:https://idss-stage.mit.edu/calendar/lids-seminar-series-jose-m-f-moura/
LOCATION:32-141\, United States
CATEGORIES:LIDS Seminar Series
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20180418T140000
DTEND;TZID=America/New_York:20180418T150000
DTSTAMP:20260524T190854
CREATED:20180223T172740Z
LAST-MODIFIED:20180223T172740Z
UID:7439-1524060000-1524063600@idss-stage.mit.edu
SUMMARY:Community-based and Peer-to-peer Electricity Markets
DESCRIPTION:Abstract \nThe deployment of distributed renewable generation capacities\, new ICT capabilities\, as well as a more proactive role of consumers\, are all motivating rethinking electricity markets in a more distributed and consumer-centric fashion. After motivating the design of various forms of consumer-centric electricity markets\, we will focus on two alternative constructs (which could actually be unified) consisting in community-based and peer-to-peer electricity markets. The mathematical framework for these markets will be described\, with focus on negotiation and clearing algorithms in a distributed and decentralized setup. Opportunities and challenges related to these markets\, both mathematical and related to real-world applications\, will be discussed. Especially\, we will look at fairness aspects\, product differentiation\, as well as the design of network charges to account for ‘actual’ usage of a network. \nBiography \nPierre Pinson is a Professor at the Centre for Electric Power and Energy (CEE) of the Technical University of Denmark (DTU\, Dept. of Electrical Engineering)\, also heading a group focusing on Energy Analytics & Markets. He holds an M.Sc. In Applied Mathematics from INSA Toulouse and a Ph.D. In Energy Engineering from Ecole de Mines de Paris (France). He acts (or has acted) as an Editor for the IEEE Transactions on Power Systems\, the International Journal of Forecasting and Wind Energy. His main research interests are centered around the proposal and application of mathematical methods for electricity markets and power systems operations\, including forecasting. He has published extensively in some of the leading journals in Meteorology\, Power Systems Engineering\, Statistics and Operations Research. He has been a visiting researcher at the University of Oxford (Mathematical Institute) and the University of Washington in Seattle (Dpt. of Statistics)\, as well as a scientist at the European Center for Medium-range Weather Forecasts (ECMWF\, UK) and a visiting professor at Ecole Normale Superieure (Rennes\, France). In 2019 he will be a Simons Fellow at the University of Cambridge\, Isaac Newton Institute (“The mathematics of energy systems”). He is leading a number of initiatives aiming to profundly rethink electricity markets for future renewable-based power systems and with a more proactive role of consumers. This focus on consumer-centric and community-driven electricity markets translates into proposals for peer-to-peer energy exchange\, from mathematical framework to actual demonstration in Denmark.
URL:https://idss-stage.mit.edu/calendar/community-based-and-peer-to-peer-electricity-markets/
LOCATION:MIT Building E18\, Room 304\, Ford Building (E18)\, 50 Ames Street\, Cambridge\, MA\, United States
CATEGORIES:LIDS Seminar Series
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20180410T150000
DTEND;TZID=America/New_York:20180410T160000
DTSTAMP:20260524T190854
CREATED:20180223T172617Z
LAST-MODIFIED:20180223T172617Z
UID:7437-1523372400-1523376000@idss-stage.mit.edu
SUMMARY:Finding Online Extremists in Social Networks
DESCRIPTION:Abstract \nOnline extremists in social networks pose a new form of threat to the general public. These extremists range from cyber bullies who harass innocent users to terrorist organizations such as ISIS that use social networks to spread propaganda. Currently\, social networks suspend the accounts of such extremists in response to user complaints\, but these extremist users simply create new accounts and continue their activities. In this talk\, we present a new set of operational capabilities to help authorities mitigate the threat posed by online extremist groups in social networks. \nUsing data from several hundred thousand extremist accounts on Twitter\, we develop a behavioral model for these users\, in particular\, what their accounts look like and who they connect with. This model is used to identify new extremist accounts by predicting if they will be suspended for extremist activity. We also use this model to track existing extremist users as they create new accounts by identifying if two accounts belong to the same user. Finally\, we use this model as the basis for an efficient policy to search the social network for suspended users’ new accounts. Our search approach is based on a variant of the classic Polya’s urn setup. We find a simple characterization of the optimal search policy for this model under fairly general conditions. Our search policy and main theoretical results generalize easily to search problems in other fields. \nJoint work with Jytte Klausen and Christopher Marks. \nBiography \nTauhid is an Assistant Professor of Operations Management at the MIT Sloan School of Management. He received his BS\, MEng\, and Ph.D. degrees in electrical engineering and computer science from MIT. His research focuses on solving operational problems involving social network data using probabilistic models\, network algorithms\, and modern statistical methods. Some of the topics he studies in the social networks space include predicting the popularity of content\, finding online extremists\, and geo-locating users. His broader interests cover data-driven approaches to investing in startup companies\, non-traditional choice modeling\, algorithmic sports betting\, and biometric data. His work has been featured in the Wallstreet Journal\, Wired\, Mashable\, the LA Times\, and Time Magazine.
URL:https://idss-stage.mit.edu/calendar/finding-online-extremists-in-social-networks/
LOCATION:32-141\, United States
CATEGORIES:LIDS Seminar Series
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20180320T150000
DTEND;TZID=America/New_York:20180320T160000
DTSTAMP:20260524T190854
CREATED:20180223T172446Z
LAST-MODIFIED:20180223T172446Z
UID:7435-1521558000-1521561600@idss-stage.mit.edu
SUMMARY:LIDS Seminar Series - Lizhong Zheng
DESCRIPTION:
URL:https://idss-stage.mit.edu/calendar/lids-seminar-series-lizhong-zheng/
LOCATION:32-141\, United States
CATEGORIES:LIDS Seminar Series
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20180313T150000
DTEND;TZID=America/New_York:20180313T160000
DTSTAMP:20260524T190854
CREATED:20180223T172349Z
LAST-MODIFIED:20180223T172349Z
UID:7433-1520953200-1520956800@idss-stage.mit.edu
SUMMARY:The Power of Multiple Samples in Generative Adversarial Networks
DESCRIPTION:Abstract \nWe bring the tools from Blackwell’s seminal result on comparing two stochastic experiments from 1953\, to shine a new light on a modern application of great interest: Generative Adversarial Networks (GAN). Binary hypothesis testing is at the center of training GANs\, where a trained neural network (called a critic) determines whether a given sample is from the real data or the generated (fake) data. By jointly training the generator and the critic\, the hope is that eventually\, the trained generator will generate realistic samples. One of the major challenges in GAN is known as “mode collapse”; the lack of diversity in the samples generated by thus trained generators. We propose a new training framework\, where the critic is fed with multiple samples jointly (which we call packing)\, as opposed to each sample separately as done in standard GAN training. With this simple but fundamental departure from existing GANs\, experimental results show that the diversity of the generated samples improve significantly. We analyze this practical gain by first providing a formal mathematical definition of mode collapse and making a fundamental connection between the idea of packing and the intensity of mode collapse. Precisely\, we show that the packed critic naturally penalizes mode collapse\, thus encouraging generators with less mode collapse. The analyses critically rely on operational interpretation of hypothesis testing and corresponding data processing inequalities\, which lead to sharp analyses with simple proofs. For this talk\, Prof. Sewoong Oh will assume no prior background on GANs. \nBiography \nSewoong Oh is an Assistant Professor of Industrial and Enterprise Systems Engineering at UIUC. He received his Ph.D. from the Department of Electrical Engineering at Stanford University. Following his Ph.D.\, he worked as a postdoctoral researcher at Laboratory for Information and Decision Systems (LIDS) at MIT. His research interest is in theoretical machine learning\, including spectral methods\, ranking\, crowdsourcing\, estimation of information measures\, differential privacy\, and generative adversarial networks. He was co-awarded the best paper award at the SIGMETRICS in 2015\, NSF CAREER award in 2016 and GOOGLE Faculty Research Award.
URL:https://idss-stage.mit.edu/calendar/the-power-of-multiple-samples-in-generative-adversarial-networks/
LOCATION:32-141\, United States
CATEGORIES:LIDS Seminar Series
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20180227T150000
DTEND;TZID=America/New_York:20180227T160000
DTSTAMP:20260524T190854
CREATED:20180223T172101Z
LAST-MODIFIED:20180223T172101Z
UID:7430-1519743600-1519747200@idss-stage.mit.edu
SUMMARY:Safe Learning in Robotics
DESCRIPTION:Abstract \nA great deal of research in recent years has focused on robot learning. In many applications\, guarantees that specifications are satisfied throughout the learning process are paramount. For the safety specification\, we present a controller synthesis technique based on the computation of reachable sets using optimal control. We show recent results in system decomposition to speed up this computation\, and how offline computation may be used in online applications. We then present a method combining reachability with machine learning\, which uses approximate knowledge of the dynamics to provide a least-restrictive\, safety-preserving control law which intervenes only when the computed safety guarantees require it\, or when confidence in the computed guarantee decays in light of new observations. We will illustrate these methods on a quadrotor UAV experimental platform which we have at Berkeley. \nBiography \nClaire Tomlin is the Charles A. Desoer Professor of Engineering in EECS at Berkeley. She was an Assistant\, Associate\, and Full Professor in Aeronautics and Astronautics at Stanford from 1998 to 2007\, and in 2005 joined Berkeley. Claire works in the area of control theory and hybrid systems\, with applications to air traffic management\, UAV systems\, energy\, robotics\, and systems biology. She is a MacArthur Foundation Fellow (2006)\, an IEEE Fellow (2010)\, and in 2017 was awarded the IEEE Transportation Technologies Award.​​
URL:https://idss-stage.mit.edu/calendar/safe-learning-in-robotics/
LOCATION:32-141\, United States
CATEGORIES:LIDS Seminar Series
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20180220T150000
DTEND;TZID=America/New_York:20180220T160000
DTSTAMP:20260524T190854
CREATED:20180223T171917Z
LAST-MODIFIED:20180223T171917Z
UID:7427-1519138800-1519142400@idss-stage.mit.edu
SUMMARY:Submodular Optimization: From Discrete to Continuous and Back
DESCRIPTION:Abstract \nMany procedures in statistics and artificial intelligence require solving non-convex problems. Historically\, the focus has been to convexify the non-convex objectives. In recent years\, however\, there has been significant progress to optimize non-convex functions directly. This direct approach has led to provably good guarantees for specific problem instances such as latent variable models\, non-negative matrix factorization\, robust PCA\, matrix completion\, etc. Unfortunately\, there is no free lunch and it is well known that in general finding the global optimum of a non-convex optimization problem is NP-hard. This computational barrier has mainly shifted the goal of non-convex optimization towards two directions: a) finding an approximate local minimum by avoiding saddle points or b) characterizing general conditions under which the underlying non-convex optimization is tractable. \nIn this talk\, I will consider a broad class of non-convex optimization problems that possess special combinatorial structures. More specifically\, I will focus on maximization of stochastic continuous submodular functions that demonstrate diminishing returns. Despite the apparent lack of convexity in such functions\, we will see that first order methods can indeed provide strong approximation guarantees. In particular\, for monotone and continuous submodular functions\, we will show that projected stochastic gradient methods achieve a ½ approximation ratio. We then see how we can reach the tight (1-1/e) approximation guarantee by developing a new class of stochastic projection-free gradient methods. A simple variant of these algorithms also achieves a (1/e) approximation ratio in the non-monotone case. Finally\, by using stochastic continuous optimization as an interface\, we will also provide tight approximation guarantees for maximizing a (monotone or non-monotone) stochastic submodular set function subject to a general matroid constraint. \nIn this talk\, I will not assume any particular background on submodularity or optimization and will try to motivate and define all the necessary concepts. \nBiography \nAmin Karbasi is an assistant professor in the School of Engineering and Applied Science (SEAS) at Yale University\, where he leads the Inference\, Information\, and Decision (I.I.D.) Systems Group. Prior to that he was a post-doctoral scholar at ETH Zurich\, Switzerland (2013-2014). He obtained his Ph.D. (2012) and M.Sc. (2007) in computer and communication sciences from EPFL\, Switzerland and his B.Sc. (2004) in electrical engineering from the same university.
URL:https://idss-stage.mit.edu/calendar/submodular-optimization-from-discrete-to-continuous-and-back/
LOCATION:34-101
CATEGORIES:LIDS Seminar Series
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20180213T150000
DTEND;TZID=America/New_York:20180213T160000
DTSTAMP:20260524T190854
CREATED:20180223T171528Z
LAST-MODIFIED:20180223T171528Z
UID:7424-1518534000-1518537600@idss-stage.mit.edu
SUMMARY:Supervisory Control of Discrete Event Systems: A Retrospective and Two Recent Results on Security and Privacy
DESCRIPTION:Abstract \nLafortune will begin with a brief retrospective of the theory of supervisory control of discrete event systems\, initiated in the seminal work of Ramadge & Wonham over 30 years ago\, and compare it with recent work in formal methods in control. He will then present results from his group on two problems: (i) sensor deception attacks in the supervisory control layer of a cyber-physical system; and (ii) obfuscation of system secrets by insertion of fictitious events in the output stream of the system. In each case\, he will describe the group’s solution procedure\, which is based on synthesizing a discrete game structure that embeds all valid solutions. \nBiography \nStéphane Lafortune is a professor in the Department of Electrical Engineering and Computer Science at the University of Michigan\, Ann Arbor\, USA. He obtained his degrees from École Polytechnique de Montréal (B.Eng)\, McGill University (M.Eng)\, and the University of California at Berkeley (PhD)\, all in electrical engineering. He is a Fellow of IEEE (1999) and of IFAC (2017). \nLafortune’s research interests are in discrete event systems and include multiple problem domains: modeling\, diagnosis\, control\, optimization\, and applications to computer and software systems. He co-authored\, with C. Cassandras\, the textbook Introduction to Discrete Event Systems (2nd Edition\, Springer\, 2008). He has served as Editor-in-Chief of the journal Discrete Event Dynamic Systems: Theory and Applications since 2015.
URL:https://idss-stage.mit.edu/calendar/supervisory-control-of-discrete-event-systems-a-retrospective-and-two-recent-results-on-security-and-privacy/
LOCATION:32-141\, United States
CATEGORIES:LIDS Seminar Series
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20171205T160000
DTEND;TZID=America/New_York:20171205T170000
DTSTAMP:20260524T190854
CREATED:20171002T160334Z
LAST-MODIFIED:20190501T144332Z
UID:6543-1512489600-1512493200@idss-stage.mit.edu
SUMMARY:Regularized Nonlinear Acceleration
DESCRIPTION:We describe a convergence acceleration technique for generic optimization problems. Our scheme computes estimates of the optimum from a nonlinear average of the iterates produced by any optimization method. The weights in this average are computed via a simple linear system\, whose solution can be updated online. This acceleration scheme runs in parallel to the base algorithm\, providing improved estimates of the solution on the fly\, while the original optimization method is running. Numerical experiments are detailed on classical classification problems. \nBio: After dual PhDs from Ecole Polytechnique and Stanford University in optimisation and finance\, followed by a postdoc at U.C. Berkeley\, Alexandre d’Aspremont joined the faculty at Princeton University as an assistant then associate professor with joint appointments at the ORFE department and the Bendheim Center for Finance. He returned to Europe in 2011 thanks to a grant from the European Research Council and is now a research director at CNRS\, attached to Ecole Normale Supérieure in Paris. His research focuses on convex optimization and applications to machine learning\, statistics and finance. \n____________________________________ \nThe LIDS Seminar Series features distinguished speakers who provide an overview of a research area\, as well as exciting recent progress in that area. Intended for a broad audience\, seminar topics span the areas of communications\, computation\, control\, learning\, networks\, probability and statistics\, optimization\, and signal processing. 
URL:https://lids.mit.edu/news-and-events/events/regularized-nonlinear-acceleration
LOCATION:MIT Building 32\, Room 141\, The Stata Center (32-141)\, 32 Vassar Street\, Cambridge\, MA\, 02139\, United States
CATEGORIES:LIDS Seminar Series
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20171129T160000
DTEND;TZID=America/New_York:20171129T170000
DTSTAMP:20260524T190854
CREATED:20171002T155836Z
LAST-MODIFIED:20190501T144513Z
UID:6541-1511971200-1511974800@idss-stage.mit.edu
SUMMARY:Comparison Lemmas\, Non-Smooth Convex Optimization and Structured Signal Recovery
DESCRIPTION:In the past couple of decades\, non-smooth convex optimization has emerged as a powerful tool for the recovery of structured signals (sparse\, low rank\, finite constellation\, etc.) from possibly noisy measurements in a variety applications in statistics\, signal processing and machine learning. While the algorithms (basis pursuit\, LASSO\, etc.) are often fairly well established\, rigorous frameworks for the exact analysis of the performance of such methods are only just emerging. The talk will introduce and describe a fairly general theory for how to determine the performance (minimum number of measurements\, mean-square-error\, probability-of-error\, etc.) of such methods for various measurement ensembles (Gaussian\, Haar\, etc.). The framework enables one to assess the performance of these methods before actual implementation and allows one to optimally choose parameters such as regularizer coefficients\, number of measurements\, etc. The theory subsumes earlier results as special cases. It builds on an inconspicuous 1962 lemma of Slepian (for comparing Gaussian processes)\, as well as on a non-trivial generalization due to Gordon in 1988\, and produces concepts from convex geometry (such as Gaussian widths and Moreau envelopes) in a very natural way. The talk will also consider extensions to certain non-Gaussian settings and their applications in massive MIMO\, one-bit compressed sensing\, graphical LASSO and phase retrieval. \n\n\nBio: Babak Hassibi is the inaugural Mose and Lillian S. Bohn Professor of Electrical Engineering at the California Institute of Technology\, where he has been since 2001\, From 2011 to 2016 he was the Gordon M Binder/Amgen Professor of Electrical Engineering and during 2008-2015 he was Executive Officer of Electrical Engineering\, as well as Associate Director of Information Science and Technology. Prior to Caltech\, he was a Member of the Technical Staff in the Mathematical Sciences Research Center at Bell Laboratories\, Murray Hill\, NJ. He obtained his PhD degree from Stanford University in 1996 and his BS degree from the University of Tehran in 1989. His research interests span various aspects of information theory\, communications\, signal processing\, control and machine learning. He is an ISI highly cited author in Computer Science and\, among other awards\, is the recipient of the US Presidential Early Career Award for Scientists and Engineers (PECASE) and the David and Lucille Packard Fellowship in Science and Engineering \n\n____________________________________ \nThe LIDS Seminar Series features distinguished speakers who provide an overview of a research area\, as well as exciting recent progress in that area. Intended for a broad audience\, seminar topics span the areas of communications\, computation\, control\, learning\, networks\, probability and statistics\, optimization\, and signal processing. 
URL:https://lids.mit.edu/news-and-events/events/babak-hassibi-california-institute-technology
LOCATION:MIT Building 32\, Room 141\, The Stata Center (32-141)\, 32 Vassar Street\, Cambridge\, MA\, 02139\, United States
CATEGORIES:LIDS Seminar Series
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20171114T160000
DTEND;TZID=America/New_York:20171114T170000
DTSTAMP:20260524T190854
CREATED:20171002T155150Z
LAST-MODIFIED:20190501T144705Z
UID:6538-1510675200-1510678800@idss-stage.mit.edu
SUMMARY:Quantum Limits on the Information Carried by Electromagnetic Radiation
DESCRIPTION:In many practical applications information is conveyed by means of electromagnetic radiation and a natural question concerns the fundamental limits of this process. Identifying information with entropy\, one can ask about the maximum amount of entropy associated to the propagating wave. \nThe standard statistical physics approach to compute entropy is to take the logarithm of the number of possible energy states of a system. Since any continuum field can assume an uncountably infinite number of energy configurations\, the approach underlying any finite entropy calculation must also necessarily include some grouping of states together in a procedure known as coarse-graining or\, in information-theoretic parlance\, signal quantization. The problem then reduces to counting the eigenstates of the Hamiltonian of the quantum wave field. \nIn this talk\, we examine the relationship between entropy computations in a statistical physics and an information-theory context. In the latter context\, rather than attempting to directly count the number of energy eigenstates of the quantum wave field\, we constrain the geometry of the signal space and decompose the waveform into a minimum number of orthogonal basis modes. We then ask how many bits are required to represent any waveform in the space spanned by this optimal representation with a minimum quantized energy error. We show that for scalar quantization this entropy computation is completely analogous to the one for the number state channel of statistical physics\, and it has the attractive feature that the complexity of state counting is now replaced by the geometric problem of optimally covering the signal space by high-dimensional boxes\, whose size is lower bounded by quantum constraints. For bandlimited radiation in a three-dimensional space\, using this approach we can recover the Bekenstein entropy bound on the largest amount of information that can be radiated from a sphere of given radius. We also compare results with black body radiation occurring over an infinite spectrum of frequencies and along the way we provide some new results on the asymptotic dimensionality and $\epsilon$-entropy of bandlimited\, square-integrable signals. \n\n\nBio: Massimo Franceschetti received the Laurea degree (with highest honors) in computer engineering from the University of Naples\, Naples\, Italy\, in 1997\, the M.S. and Ph.D. degrees in electrical engineering from the California Institute of Technology\, Pasadena\, CA\, in 1999\, and 2003\, respectively. He is Professor of Electrical and Computer Engineering at the University of California at San Diego (UCSD). Before joining UCSD\, he was a postdoctoral scholar at the University of California at Berkeley for two years. His research interests are in physical and information-based foundations of communication and control systems. He was awarded the C. H. Wilts Prize in 2003 for best doctoral thesis in electrical engineering at Caltech\, the S.A. Schelkunoff Award in 2005 for best paper in the IEEE TRANSACTIONS ON ANTENNAS AND PROPAGATION\, a National Science Foundation (NSF) CAREER award in 2006\, an Office of Naval Research (ONR) Young Investigator Award in 2007\, the IEEE Communications Society Best Tutorial Paper Award in 2010\, and the IEEE Control theory society Ruberti young researcher award in 2012. \n\n____________________________________ \nThe LIDS Seminar Series features distinguished speakers who provide an overview of a research area\, as well as exciting recent progress in that area. Intended for a broad audience\, seminar topics span the areas of communications\, computation\, control\, learning\, networks\, probability and statistics\, optimization\, and signal processing. 
URL:https://lids.mit.edu/news-and-events/events/quantum-limits-information-carried-electromagnetic-radiation
LOCATION:MIT Building 32\, Room 141\, The Stata Center (32-141)\, 32 Vassar Street\, Cambridge\, MA\, 02139\, United States
CATEGORIES:LIDS Seminar Series
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20171031T160000
DTEND;TZID=America/New_York:20171031T170000
DTSTAMP:20260524T190854
CREATED:20171002T154828Z
LAST-MODIFIED:20190501T144833Z
UID:6535-1509465600-1509469200@idss-stage.mit.edu
SUMMARY:Structure\, Randomness and Universality
DESCRIPTION:What is the minimum possible number of vertices of a graph that contains every k-vertex graph as an induced subgraph? What is the minimum possible number of edges in a graph that contains every k-vertex graph with maximum degree 3 as a subgraph? These questions and related one were initiated by Rado in the 60s\, and received a considerable amount of attention over the years\, partly motivated by algorithmic applications. The study of the subject combines probabilistic arguments and explicit\, structured constructions. I will survey the topic focusing on a recent asymptotic solution of the first question\, where an asymptotic formula\, improving earlier estimates by several researchers\, is obtained by combining combinatorial and probabilistic arguments with group theoretic tools. \nBio: Noga Alon is a Baumritter Professor of Mathematics and Computer Science in Tel Aviv University\, Israel. He received his Ph. D. in Mathematics at the Hebrew University of Jerusalem in 1983 and had visiting positions in various research institutes including MIT\, the Institute for Advanced Study in Princeton\, IBM Almaden Research Center\, Bell Laboratories\, Bellcore and Microsoft Research. He joined Tel Aviv University in 1985\, served as the head of the School of Mathematical Sciences in 1999-2000\, and supervised about 20 PhD students. Since 2009 he is also a member of Microsoft Research\, Israel. He serves on the editorial boards of more than a dozen international technical journals and has given invited lectures in many conferences\, including plenary addresses in the 1996 European Congress of Mathematics and in the 2002 International Congress of Mathematician. He published more than five hundred research papers and one book. \n\n\nHis research interests are mainly in Combinatorics\, Graph Theory and their applications in Theoretical Computer Science. His main contributions include the study of expander graphs and their applications\, the investigation of derandomization techniques\, the foundation of streaming algorithms\, the development and applications of algebraic and probabilistic methods in Discrete Mathematics and the study of problems in Information Theory\, Combinatorial Geometry and Combinatorial Number Theory. \nHe is an ACM Fellow and an AMS Fellow\, a member of the Israel Academy of Sciences and Humanities since 1997 and of the Academia Europaea since 2008\, and received the Erdös prize in 1989\, the Feher prize in 1991\, the Polya Prize in 2000\, the Bruno Memorial Award in 2001\, the Landau Prize in 2005\, the Gödel Prize in 2005\, the Israel Prize in 2008\, the EMET Prize in 2011\, the Dijkstra Prize in 2016\, an Honorary Doctorate from ETH Zurich in 2013 and from the University of Waterloo in 2015. \n\n____________________________________ \nThe LIDS Seminar Series features distinguished speakers who provide an overview of a research area\, as well as exciting recent progress in that area. Intended for a broad audience\, seminar topics span the areas of communications\, computation\, control\, learning\, networks\, probability and statistics\, optimization\, and signal processing. 
URL:https://lids.mit.edu/news-and-events/events/joint-seminar-csail-theory-computation-toc
LOCATION:32-G449 (Kiva)\, United States
CATEGORIES:LIDS Seminar Series
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20171024T160000
DTEND;TZID=America/New_York:20171024T170000
DTSTAMP:20260524T190854
CREATED:20171002T154138Z
LAST-MODIFIED:20190501T145009Z
UID:6530-1508860800-1508864400@idss-stage.mit.edu
SUMMARY:Regularized Nonlinear Acceleration
DESCRIPTION:We describe a convergence acceleration technique for generic optimization problems. Our scheme computes estimates of the optimum from a nonlinear average of the iterates produced by any optimization method. The weights in this average are computed via a simple linear system\, whose solution can be updated online. This acceleration scheme runs in parallel to the base algorithm\, providing improved estimates of the solution on the fly\, while the original optimization method is running. Numerical experiments are detailed on classical classification problems. \nBio: After dual PhDs from Ecole Polytechnique and Stanford University in optimisation and finance\, followed by a postdoc at U.C. Berkeley\, Alexandre d’Aspremont joined the faculty at Princeton University as an assistant then associate professor with joint appointments at the ORFE department and the Bendheim Center for Finance. He returned to Europe in 2011 thanks to a grant from the European Research Council and is now a research director at CNRS\, attached to Ecole Normale Supérieure in Paris. His research focuses on convex optimization and applications to machine learning\, statistics and finance. \n\n____________________________________ \nThe LIDS Seminar Series features distinguished speakers who provide an overview of a research area\, as well as exciting recent progress in that area. Intended for a broad audience\, seminar topics span the areas of communications\, computation\, control\, learning\, networks\, probability and statistics\, optimization\, and signal processing. 
URL:https://lids.mit.edu/news-and-events/events/alexandre-tsybakov-ensae-paristech
LOCATION:MIT Building 32\, Room 141\, The Stata Center (32-141)\, 32 Vassar Street\, Cambridge\, MA\, 02139\, United States
CATEGORIES:LIDS Seminar Series
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20171017T160000
DTEND;TZID=America/New_York:20171017T170000
DTSTAMP:20260524T190854
CREATED:20171002T153935Z
LAST-MODIFIED:20190501T145140Z
UID:6528-1508256000-1508259600@idss-stage.mit.edu
SUMMARY:The Maps Inside Your Head
DESCRIPTION:How do our brains make sense of a complex and unpredictable world? In this talk\, I will discuss an information theory approach to the neural topography of information processing in the brain. First I will review the brain’s architecture\, and how neural circuits map out the sensory and cognitive worlds. Then I will describe how highly complex sensory and cognitive tasks are carried out by the cooperative action of many specialized neurons and circuits\, each of which has a simple function. I will illustrate my remarks with one sensory example and one cognitive example. For the sensory example\, I will consider the sense of smell (“olfaction”)\, whereby humans and other animals distinguish vast arrays of odor mixtures using very limited neural resources. For the cognitive example\, I will consider the “sense of place”\, that is\, how animals mentally represent their physical location. Both examples demonstrate that brains have evolved neural circuits that exploit sophisticated principles of mathematics and information processing – principles that scientists have only recently discovered. \nBio: Vijay Balasubramanian is the Cathy and Marc Lasry Professor in the Physics Department at the University of Pennsylvania\, where he is also Director of the Computational Neuroscience Initiative. He received B.Sc. degrees in Physics and Computer Science\, and an M.Sc. in Computer Science\, from MIT. He earned a Ph.D. in Theoretical Physics at Princeton University\, and was a Junior Fellow of the Harvard Society of Fellows. \n\n\n____________________________________ \n\nThe LIDS Seminar Series features distinguished speakers who provide an overview of a research area\, as well as exciting recent progress in that area. Intended for a broad audience\, seminar topics span the areas of communications\, computation\, control\, learning\, networks\, probability and statistics\, optimization\, and signal processing. 
URL:https://lids.mit.edu/news-and-events/events/maps-inside-your-head
LOCATION:MIT Building 32\, Room 141\, The Stata Center (32-141)\, 32 Vassar Street\, Cambridge\, MA\, 02139\, United States
CATEGORIES:LIDS Seminar Series
END:VEVENT
END:VCALENDAR