Vincent Y. F. Tan 陈延福 (Pronouns: he/him/his/他)
|
Associate Professor,
Affiliate,
On Sabbatical at IORA in AY 2022/23 Office: E4-06-06 (ECE) and S17-05-20 (Mathematics)
My Curriculum Vitae.
|
News (Archive)
- Advertisement: Looking to hire motivated postdocs. Please see this advertisement for more details.
- Jan 2023: Posted a paper on Probably Anytime-Safe Stochastic Combinatorial Semi-Bandits. Joint work with Ph.D. student Yunlong Hou and former Ph.D. student Zixin Zhong. In this paper, we consider the stochastic combinatorial semi-bandit problem where at each time, the agent's choices at every time instant should contain items whose sum of risks does not exceed a certain risk budget.
- Jan 2023: Paper on "Near-Optimal Learning of Tree-Structured Distributions by Chow-Liu" accepted by the SIAM Journal of Computing. Joint work with Arnab Bhattacharyya, Sutanu Gayen, Eric Price, and N. V. Vinodchandran.
- Jan 2023: Posted a paper on Codes for Correcting Asymmetric Adjacent Transpositions and Deletions. Joint work with Ph.D. student Shuche Wang and collaborator Vu Van Khu. In this paper, we design and analyze codes for correcting asymmetric adjacent transpositions and deletions.
- Jan 2023: Successfully organized the Information Theory and Data Science workshop at the Institute of Mathematical Sciences with Po-Ling Loh and Jonathan Scarlett. Thanks to all participants!
- Jan 2023: Paper on Understanding and Mitigating Dimensional Collapse in Heterogeneous Federated Learning accepted to ICLR 2023! Joint work with Ph.D. student Yujun Shi and other co-authors. In this paper, we address the data heterogeneity problem in federated learning by mitigate dimensional collapse. See you in Kigali, Rwanda.
- Jan 2023: Paper on the effect of pseudo-labeling on the generalization error of the semi-supervised Gibbs algorithm accepted to AISTATS 2023! Joint work with Haiyun He, Gholamali Aminian, Yuheng Bu and Miguel Rodrigues. We show that the generalization performance of SSL with pseudo-labeling is affected not only by the information between the output hypothesis and input training data but also by the information shared between the labeled and pseudo-labeled data samples.
- Jan 2023: Awarded an MOE AcRF Tier 1 grant "The Benefits of Active Learning and Testing of Graphical Structure". The total project value is $146,800.
- Jan 2023: Congrats to Ph.D. student Fengzhuo Zhang for receiving the SDSC Dissertation Research Fellowship 2022.
- Jan 2023: Invited to the 2023 Information Theory and Applications (ITA) Workshop. See you in San Diego.
- Dec 2022: Received 4.8/5.0 for my teaching of some tutorials of the B.Tech. module TG1401 Engineering Mathematics. See review quotes here.
- Dec 2022: Paper on establishing the asymptotic Nash equilibrium for the M-ary sequential adversarial hypothesis testing game accepted to the IEEE Transactions on Information Forensics and Security. Joint work with Ph.D. student Jiachun Pan and postdoc Yonglong Li. In this paper, we derive a pair of strategies under which the asymptotic Nash equilibrium of the sequential hypothesis testing game is attained.
- Dec 2022: Paper on best arm identification in restless Markov multi-armed bandits accepted to the IEEE Transactions on Information Theory. Joint work with postdocs P. N. Karthik and Srinivas Kota Reddy. In this work, we provide problem instance-dependent asymptotic bounds on the growth rate of the expected time required to find the index of the best arm where the arms evolve according to Markov chains.
- Nov 2022: Traveling to NeurIPS 2022. See you in New Orleans!
- Nov 2022: Posted a preprint on Minimizing the Accumulated Trajectory Error to Improve Dataset Distillation. Joint work with Ph.D. students Jiawei Du and Yidi Jiang.
- Nov 2022: Paper on "Almost Cost-Free Communication in Federated Best Arm Identification" accepted to the 37th AAAI Conference on Artificial Intelligence (AAAI), 2023 (Acceptance Rate: 1721/8777 ≈ 19.6%). Joint work with postdocs Srinivas Kota Reddy and P. N. Karthik. In this paper, we study the problem of best arm identification in a federated learning multi-armed bandit setup with a central server and multiple clients and elucidate the tradeoff between communication cost and number of arm selections.
- Nov 2022: Paper on almost optimal variance-constrained best arm identification accepted to the IEEE Transactions on Information Theory. This is joint work with Ph.D. student Yunlong Hou and former student Zixin Zhong. In this paper, we consider the BAI problem under a stringent variance constraint. We propose an algorithm VA-LUCB and show that its performance meets an information-theoretic lower bound.
- Nov 2022: Paper on active learning of homogeneous Ising trees accepted to the IEEE Transactions on Information Theory. This is joint work with Ph.D. student Fengzhuo Zhang and former postdoc Anshoo Tandon. In this work, we use active learning techniques to improve the learnability of homogeneous Ising tree models. We show that for high edge correlations, the error exponent can be improved by at least 40% over its passive counterpart.
- Nov 2022: Posted a monograph on Common Information, Noise Stability and Their Extensions on the arXiv. Joint work with Lei Yu.
- Nov 2022: Technical Program Committee (TPC) Co-Chair of the 2025 International Symposium on Information Theory (ISIT) in Ann Arbor, Michigan.
Research Interests and Open Positions
Information Theory
Machine Learning
Statistical Signal Processing
I am actively recruiting graduate students in the areas of my research interest above. There are various channels to be admitted to NUS including to the Faculty of Engineering, the Faculty of Science, the Institute of Operations Research and Analytics, the Institute of Data Science, or the Asian Institute of Digital Finance.
There are also multiple positions for talented postdoctoral scholars. Postdoctoral scholars with strong publication records and showing interest in the above research topics are also encouraged to contact me to check with me if there are available positions. Please see this advertisement as well as this.
Selected Recent Publications
Towards Understanding and Mitigating Dimensional Collapse in Heterogeneous Federated Learning
Yujun Shi, Jian Liang, Wenqing Zhang, Vincent Y. F. Tan, and Song Bai
Proc. of the 11th International Conference on Learning Representations (ICLR), Kigali, Rwanda, May 2023 (AR ≈ 31.8%)How Does Pseudo-Labeling Affect the Generalization Error of the Semi-Supervised Gibbs Algorithm?
Haiyun He, Gholamali Aminian, Yuheng Bu, Miguel R. D. Rodrigues, and Vincent Y. F. Tan
Proc. of 26th International Conference on Artificial Intelligence and Statistics (AISTATS), Valencia, Spain, Apr 2023 (AR ≈ 29%)Almost Cost-Free Communication in Federated Best Arm Identification
Kota Srinivas Reddy, P. N. Karthik, and Vincent Y. F. Tan
Proc. of the 37th AAAI Conference on Artificial Intelligence (AAAI), Washington DC, USA, Feb 2023 (AR: 1721/8777 ≈ 19.6%)Best-Arm Identification in Restless Markov Multi-Armed Bandits [Slides]
P. N. Karthik, Kota Srinivas Reddy, and Vincent Y. F. Tan
IEEE Transactions on Information Theory, 2022+Fast Beam Alignment via Pure Exploration in Multi-armed Bandits [Slides] [Code]
Yi Wei, Zixin Zhong, and Vincent Y. F. Tan
IEEE Transactions on Wireless Communications, 2022+
Oldies But Goldies
Information Spectrum Approach to Strong Converse Theorems for Degraded Wiretap Channels
Vincent Y. F. Tan and Matthieu R. Bloch
IEEE Transactions on Information Forensics and Security, Vol. 10, No. 9, Pages 1891 - 1904, Sep 2015Asymptotic Estimates in Information Theory with Non-Vanishing Error Probabilities
Vincent Y. F. Tan
Foundations and Trends® in Communications and Information Theory, Vol. 11, Nos. 1-2, Pages 1 – 184, 2014Automatic Relevance Determination in Nonnegative Matrix Factorization with the β-Divergence [Slides] [Code]
Vincent Y. F. Tan and Cédric Févotte
IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 35, No. 7, Pages 1592 - 1605, Jul 2013A Large-Deviation Analysis of the Maximum-Likelihood Learning of Markov Tree Structures [Slides]
Vincent Y. F. Tan, Animashree Anandkumar, Lang Tong and Alan S. Willsky
IEEE Transactions on Information Theory, Vol. 57, No. 3, Pages 1714 - 1735, Mar 2011Estimating Signals with Finite Rate of Innovation from Noisy Samples: A Stochastic Algorithm [Code]
Vincent Y. F. Tan and Vivek K Goyal
IEEE Transactions on Signal Processing, Vol. 56, Issue 10, No. 5135 - 5145, Oct 2008