Short courses, tutorials, and invited talks

Short courses

• 5H mini-course on “Random Matrix Theory for Machine Learning”, within thematic trimester on “Beyond classical regimes in statistical inference and machine learning” Centre International de Mathématiques et Informatique de Toulouse (CIMI), Toulouse, France, 2024.
• 12H mini-course on “Random Matrix Theory for Modern Machine Learning: New Intuitions, Improved Methods, and Beyond” at Institut de Recherche en Informatique de Toulouse (IRIT), Toulouse, France, July, 2024. See slides of Part 1, Part 2, Part 3, and Part 4.
• 2H short course at SDS, Fudan University, see slides
• 5H short course on “Random Matrix Theory and Its Applications in ML” at Jiangsu Normal University, China, 2024. Links to recordings: part 1 on RMT and part 2 on its application to deep learning. See slides of part 1 and part 2
• 3H mini-course on “Random Matrix Theory in Deep Learning: An Introduction”, Northeast Normal University, Changchun, Nov, 2023. See slides

Tutorials

Invited talks

• Invited talk on “Recent Advances in Random Matrix Methods for Deep Learning Theory” at CSML 2024, Shanghai, China, August, 2024. See slides
• Invited talk on “A Random Matrix Approach to Explicit and Implicit Deep Neural Networks” at Institut de Mathématiques de Toulouse (IMT), Toulouse, France, July, 2024. See slides
• Invited talk on “Neural Tangent Kernel in High Dimensions: From Theory to Practice”, Beijing Jiaotong University School of Computer and Information Technology, Beijing, Oct, 2023. See slides
• Invited talk on “Random Matrix Methods for Machine Learning: Lossless Compression of Large Neural Networks” at The China conference on Scientific Machine Learning (CSML 2022), see slides here.
• I will be talking about the interface between RMT and ML at School of Physical & Mathematical Sciences, Nanyang Technological University (NTU), see slides here.
• I will be talking about some recent work on the interaction between RMT and machine learning at the SDS Workshop on “Topics in Random Matrix Theory” at CUHK-Shenzhen, see more details here!
• Invited talk on “A Data-dependent Theory of Overparameterization: Phase Transition, Double Descent, and Beyond” at Workshop on the Theory of Over-parameterized Machine Learning (TOPML) 2021. See slides, two-page abstract, and paper.
• Invited talk on “Performance-complexity Trade-off in Large Dimensional Spectral Clustering” at Statistics Seminar, Research School of Finance, Actuarial Studies and Statistics, Australian National University, Canberra, March, 2021. See slides here.
• Invited talk on “Performance-complexity trade-off in large dimensional spectral clustering” at STA 290 Seminar, Department of Statistics, University of California, Davis, January, 2021. See slides here.
• Invited talk on “Dynamical aspects of learning linear neural networks” at Second Symposium on Machine Learning and Dynamical Systems, Fields Institute, Toronto, September, 2020. See slides here.
• Invited talk on “Random matrix advances in large dimensional machine learning” at Random Matrices and Complex Data Analysis Workshop, Shanghai University of Finance and Economics, Shanghai, China, 10-12 December, 2019. See slides here.
• Invited talk on “RMT understanding for machine learning” at Random Matrix Theory: Applications in the Information Era conference, Kraków, Poland, 29 April - 2 May, 2019, see slides here.
• Invited talk on “RMT viewpoint of learning with gradient descent” at DIMACS workshop on Randomized Numerical Linear Algebra, Statistics, and Optimization, Rutgers University, USA, 16-18 September, 2019. See slides here.