talks

Below are some slides from a selection of talks (research and other). There may be some small errors/typos in a few of the slides: the most recent versions of the talk should have fewer typos but YMMV. If you see something egregious let me know and I’ll try to update!

Selected research talks

• [2025-08-15] Are modern ML models like scientific instruments?, Monash University (Melbourne, Australia)

• [2025-08-13] Communication against restricted adversaries: between Shannon and Hamming, University of Melbourne (Melbourne, Austalia)

• [2025-08-07] Communication against restricted adversaries: between Shannon and Hamming, University of Sydney (Sydney, Australia)

• [2025-08-04] An information theorist’s tour of differential privacy, Macquarie University (Sydney, Australia)

• [2025-07-31] Flexible Tensor Decompositions for Learning and Optimization, Southwest Jiaotong University (Chengdu, China)

• [2025-07-28] Are modern ML models like scientific instruments?, AICIT 2025 (Guangzhou, China)

• [2024-07-24] Communication against restricted adversaries: between Shannon and Hamming, Chinese University of Hong Kong (CUHK) (Hong Kong SAR, China)

• [2025-05-20] An information theorist visits differential privacy, INFORMED-AI Seminar, University of Bristol (Bristol, UK)

• [2025-04-07] Flexible Tensor Decompositions for Learning and Optimization, IIIT-Hyderabad and IIT-Hyderabad (Hyderabad, India)

• [2025-03-28] Flexible Tensor Decompositions for Learning and Optimization, Center for Information and Systems Engineering, Boston University, (Boston, USA)

• [2025-01-16] Exploring strange neu(ral network) worlds with well-worn tools, BIRS Workshop on Machine Learning and Statistics: From Theory to Practice, Chennai Mathematical Institute, (Chennai, India)

• [2024-07-18] Learning with Structured Tensor Decompositions, IEEE ITSOC Distinguished Lecture, Center for Advanced Mathematical Sciences, American University of Beirut, (Beirut, Lebanon)

• [2024-07-04] Computationally Efficient Codes for Adversarial Binary-Erasure Channels, Imperial College, (London, UK)

• [2024-04-15] Differential privacy can also enable scientific collaboration, IEEE ITSOC Distinguished Lecture, Sungkyunkwan University, (Suwon, South Korea)

Posters

• [2025-04-08] Y. Tao, A.D. Sarwate, Differentially Private Distribution Estimation Using Functional Approximation, IEEE ICASSP 2025, Hyderabad, India

Other presentations

• [2024-09-19] Getting Into Undergraduate Research in ECE at Rutgers, Undergraduate Research Discovery Day, Rutgers University