Advances in Differential Privacy: Getting More for Less

Cynthia Dwork (Harvard University, USA)

Abstract:

Differential privacy is a definition of privacy tailored to the analysis of large datasets.  The key to the success of differential privacy is the ability to quantify and reason about cumulative privacy loss over many differentially private interactions. When upper bounds on privacy loss are loose, the deployment of the algorithms is by definition conservative, “leaving something on the table.” Under high levels of composition, much potential utility is lost. We survey two general approaches to getting more utility: privacy amplification techniques, which are algorithmic, and definitional changes, which lead to tighter analyses of existing algorithms.

Biography:

Cynthia Dwork, Gordon McKay Professor of Computer Science at the Harvard Paulson School of Engineering, Radcliffe Alumnae Professor at the Radcliffe Institute for Advanced Study, and Affiliated Faculty at Harvard Law School and Department of Statistics, uses theoretical computer science to place societal problems on a firm mathematical foundation.

She was awarded the Edsger W. Dijkstra Prize in 2007 in recognition of some of her earliest work establishing the pillars on which every fault tolerant system has been built for a generation (Dwork, Lynch, and Stockmeyer, 1984).

Her contributions to cryptography include the launching of non-malleable cryptography, the subfield of modern cryptography that studies — and remedies — the failures of cryptographic protocols to compose securely (Dolev, Dwork, and Naor, 1991).  She is a co-inventor of the first public-key cryptosystem based on lattices, the current best bet for cryptographic constructions that will remain secure even against quantum computers (Ajtai and Dwork, 1997). More recently, Dwork spearheaded a successful effort to place privacy-preserving analysis of data on a firm mathematical foundation.  A cornerstone of this effort is the invention of Differential Privacy (Dwork, McSherry, Nissim, and Smith, 2006, Dwork 2006), now the subject of intense activity across many disciplines and recipient of the Theory of Cryptography Conference 2016 Test-of-Time award and the 2016 Gödel Prize. Now widely used in industry – for example by Google, MIcrosoft, Uber, and, most prominently, by Apple – differential privacy will also be the foundation of the Disclosure Avoidance System in the 2020 US Decennial Census.

Dwork was educated at Princeton and Cornell.  She received her BSE (with honors) in electrical engineering and computer science at Princeton University, where she also received the Charles Ira Young Award for Excellence in Independent Research, the first woman ever to do so.  She received her M.Sc. and Ph.D. degrees in computer science at Cornell University.

Dwork is a member of the US National Academy of Sciences and the US National Academy of Engineering, and is a fellow of the ACM, the American Academy of Arts and Sciences, and the American Philosophical Society.

A (Con)Sequential View of Information for Statistical Learning and Optimization

Tara Javidi (University of California, San Diego)

This talk, in contrast, highlights important and challenging problems in machine learning, optimization, statistics, and control theory, where the problem of acquiring information in an adaptive manner arises very naturally. Thus, I will argue that an increased emphasis on (teaching) feedback information theory can provide vast and exciting research opportunities at the intersection of information theory and these fields. In particular, I will revisit simple-to-teach results in feedback information theory including sequential hypothesis testing, arithmetic coding, successive refinement, noisy binary search, and posterior matching. I will also highlight the successful application of these sequential techniques in a variety of problem instances such as black-box optimization, distribution estimation, and active machine learning with imperfect labels.

Biography:

Tara Javidi received her BS in electrical engineering at Sharif University of Technology, Tehran, Iran. She received her MS degrees in electrical engineering (systems) and in applied mathematics (stochastic analysis) from the University of Michigan, Ann Arbor as well as her Ph.D. in electrical engineering and computer science in 2002. From 2002 to 2004, Tara Javidi was an assistant professor at the Electrical Engineering Department, University of Washington, Seattle. In 2005, she joined the University of California, San Diego, where she is currently a professor of electrical and computer engineering and a founding co-director of the Center for Machine-Intelligence, Computing and Security.

Tara Javidi’s research interests are in theory of active learning, information acquisition and statistical inference, information theory with feedback,  stochastic control theory, and wireless communications and communication networks. Tara served as a Distinguished Lecturer of the IEEE Information Theory Society (2017/18) as well as  Communications Society (2019/20). She is also a member of the Board of Governors of the IEEE Information Theory Society (2017/18/19-2020/21/22).

Tara Javidi is a Fellow of IEEE. She and her Phd students are recipients of the 2021 IEEE Communications Society & Information Theory Society Joint Paper Award. She also received the 2018 and 2019 Qualcomm Faculty Award for her contributions to wireless technology. Tara Javidi was a recipient of the National Science Foundation early career award (CAREER) in 2004, Barbour Graduate Scholarship, University of Michigan, in 1999, and the Presidential and Ministerial Recognitions for Excellence in the National Entrance Exam, Iran, in 1992. At UCSD, she has also received awards for her exceptional University service/leadership and contributions to diversity.

Listen to the noise

Muriel Médard (Massachusetts Institute of Technology, USA)

Abstract: 

Biography:

Coded Caching: Research Trends and Challenges

Balaji Sundar Rajan (Indian Institute of Science)

Abstract:

Coded caching deals with reducing the peak hour load in a wireless network by proper choice of placement of data apriori in the caches accessible to the users and reducing the delivery load by coding when the demands of the users are made known to the server later. Starting from the seminal work by Maddah-Ali and Niesen, results have been reported with several variations like, centralized and decentralized placement, multi-antenna server, coded vs uncoded placement, shared link and D2D network, unequal cache size, data and demand privacy, etc. In this talk, after a quick survey of key results, recent research trends will be highlighted along with the challenges encountered. 

Biography:

Balaji Sundar Rajan received his B.Sc. degree in mathematics from Madras University, B.Tech. degree in electronics from Madras Institute of Technology, India, and M.Tech. and Ph.D. degrees in electrical engineering from Indian Institute of Technology, Kanpur, India. He was a faculty member at the Department of Electrical Engineering, Indian Institute of Technology, Delhi, from 1990 to 1997. He has been a Professor with the Department of Electrical Communication Engineering at Indian Institute of Science, Bangalore, since 1998.

Sundar Rajan was an Associate Editor of IEEE Transactions On Information Theory (2008-2011 and 2013-2015), an Editor of IEEE Transactions on Wireless Communications (2007-2011) and an Editor of IEEE Wireless Communications Letters (2012-2015). He served as a Technical Program Co-Chair of the IEEE Information Theory Workshop (ITW’02), held in Bangalore, in 2002 and was a member of Fellow Evaluation Standing Committee of IEEE Communications Society (2019-2021). He is a fellow of IEEE, the Indian National Academy of Engineering, the Indian National Science Academy, the Indian Academy of Sciences, and the National Academy of Sciences, India. He is a recipient of Prof. Rustum Choksi Award by Indian Institute of Science for Excellence in Research in Engineering for the year 2009, the IETE Pune Center’s S.V.C Aiya Award for Telecom Education in 2004, and Best Academic Paper Award at the IEEE WCNC 2011.

Sundar Rajan’s current research interests include Coded Caching, Private Information Retrieval (PIR), Private Information Delivery, Network Coding and Coding for MIMO and multi-user communication.