2021 Theorinet Retreat

/ September 23, 2021/ Uncategorized


September 20th, 2021: Session 1: Overparameterization

September 20th, 2021 Session 2: Optimization and robustness

September 21st, 2021: Session 1: Learning on graphs

September 21st, 2021: Session 2: Limits of learning

TimeSpeakerAffiliationTitle of Presentation

Monday September 20

012:00-12:30 PM ESTYaodong YuUC BerkeleyUnderstanding Generalization in Adversarial Training via the Bias-Variance Decomposition
012:30-1:00 PM ESTSalma TarmounJohns Hopkins UniversityUnderstanding the Dynamics of Gradient Flow in Overparameterized Linear Models
01:00-1:30 PM ESTHancheng MinJohns Hopkins UniversityOn the convergence and implicit bias of overparametrized linear networks
01:30-2:00 PM ESTTeresa HuangJohns Hopkins UniversityDimensionality reduction in overparameterized regression
02:00-2:30 PM ESTMeena JagadeesanUC BerkeleyInductive Bias of Multi-Channel Linear Convolutional Networks with Bounded Weight Norm
Optimization and robustness
03:30-4:00 PM ESTYaodong YuUC BerkeleyReduNet: A White-box Deep Network from the Principle of Maximizing Rate Reduction
04:00-4:30 PM ESTChenwei WuDuke UniversityGuarantees for Tuning the Step Size using a Learning-to-Learn Approach
04:30-5:00 PM ESTAlexander RobeyUniversity of PennsylvaniaModel-Based Robust Deep Learning
05:00-5:30 PM ESTChinmay MaheshwariUC BerkeleyZeroth-Order Methods for Convex-Concave Minmax Problems: Applications to Decision-Dependent Risk Minimization

Tuesday September 21

Learning on graphs
012:00-12:30 PM ESTSohir MaskeyUniversity of MunichTransferability of Graph Neural Networks
012:30-1:00 PM ESTLuana RuizUniversity of PennsylvaniaLarge-Scale Graph Information Processing
01:00-1:30 PM ESTAlejandro ParadaUniversity of PennsylvaniaLarge-Scale Graph Information Processing
01:30-2:00 PM ESTHans RiessUniversity of PennsylvaniaGNNs & the Tarski Laplacian
Limits of learning
03:00-3:30 PM ESTJayanta DeyJohns Hopkins UniversityLifelong Learning: Theory and Practice
03:30-4:00 PM ESTAnastasios TsiamisUniversity of PennsylvaniaLinear Systems can be Hard to Learn
04:00-4:30 PM ESTLihua LeiStanford UniversityDistribution-Free, Risk-Controlling Prediction Sets
04:30-5:00 PM ESTNatalia MartinezDuke UniversityBlind Pareto Fairness and Subgroup Robustness

Share this Post