Quanquan Gu

Recent Talks

• On the Convergence of Monte Carlo Methods with Stochastic Gradients
  Simons Institute Workshop on Sampling Algorithms and Geometries on Probability Distributions, October, 2021.[slides]

• Stochastic Variance-Reduced High-order Optimization for Nonconvex Optimization
  ICML 2021 Workshop on Beyond first order methods in machine learning systems, July, 2021.[slides]

• On the Implicit Bias of Stochastic Gradient Descent with Moderate Learning Rate
  UCSD Halıcıoğlu Data Science Institute Seminar, April, 2021.[slides]

• Stochastic Variance-Reduced Cubic Regularized Newton Methods for Nonconvex Optimization
  SIAM CSE Mini-symposium on beyond first-order algorithms in modern machine learning systems, March, 2021.[slides]

• Understanding Overparameterized Deep Neural Networks: From Optimization To Generalization
  IJCAI Early Career Talk, January 2021.[slides]

• Learning Wide Neural Networks: Polylogarithmic Over-parameterization and A Mean Field Perspective
  Northwesern University IDEAL Theory of Deep Learning Seminar, October, 2020. [slides]

• Epidemic Model Guided Machine Learning for COVID-19 Forecasts
  ICLR 2021 Workshop on Machine Learning for Preventing and Combating Pandemics, May, 2021. [slides]
  UCSB Second Annual Responsible Machine Learning Summit, October, 2020. [slides]
  D. E. Shaw Technical Talk Forum, June, 2020. [slides]
  Institute for Digital Research and Education (IDRE), June, 2020. [slides]
  UCLA Computer Science Department Seminar, May, 2020. [slides]
  AI for COVID-19 in LA Symposium, May, 2020. [slides]

• Understanding, Improving and Evaluating Adversarial Robustness in Deep Learning
  Johns Hopkins University Machine Learning Seminar, September, 2020. [slides]
  KDD 2020 Workshop on Adversarial Learning Methods for Machine Learning and Data Mining, August, 2020. [slides]

• A Generalized Neural Tangent Kernel Analysis for Two-layer Neural Networks
  Simons Institute Deep Learning Reunion Workshop, August, 2020. [slides]

• Learning Over-parameterized Neural Networks: From Neural Tangent Kernel to Mean-field Analysis
  IPAM Workshop on PDE and Inverse Problem Methods in Machine Learning, April, 2020. [slides]
  UCSD AI Seminar, February, 2020. [slides]

• On the Optimization and Generalization of Neural Networks: A Mean-Field Perspective
  Information Theory and Applications Workshop, February, 2020. [slides]

• Towards Understanding Overparameterized Deep Neural Networks: From Optimization To Generalization
  TTIC Workshop on "Recent Trends in Clustering and Classification", September, 2019. [slides]
  Machine Learning Theory Workshop at Peking University, June, 2019. [slides]

• Two facets of stochastic optimization: continuous-time dynamics and discrete-time algorithms
  Workshop on "Interplay between Control, Optimization, and Machine Learning" at American Control Conference, July, 2019. [slides]

• New Variance Reduction Algorithms for Nonconvex Finite-Sum Optimization
  USC Machine Learning Seminar, November, 2018. [slides]

• Closing the Generalization Gap of Adaptive Gradient Methods in Training Deep Neural Networks
  USC Information Science Institute AI Seminar, November, 2018. [slides]

Tutorials

• Benign Overfitting: From Minimum-norm Interpolator to Stochastic Gradient Descent
  BAAI conference, May, 2021. [slides]

• Nonconvex Optimization for Knowledge Discovery and Data Mining
  SIAM Interational Conference on Data Mininng, April, 2019. [slides]