First-order optimization methods tend to inherently favor certain soluti...
The support vector machine (SVM) is a supervised learning algorithm that...
We provide a unified framework, applicable to a general family of convex...
In many bandit problems, the maximal reward achievable by a policy is of...
The classical algorithms for online learning and decision-making have th...
Data augmentation (DA) is a powerful workhorse for bolstering performanc...
We study the complexity of computing stationary Nash equilibrium (NE) in...
Overparametrized neural networks tend to perfectly fit noisy training da...
Model selection in contextual bandits is an important complementary prob...
State-of-the-art deep learning classifiers are heavily overparameterized...
The rapid recent progress in machine learning (ML) has raised a number o...
We introduce the "inverse bandit" problem of estimating the rewards of a...
The growing literature on "benign overfitting" in overparameterized mode...
We study convergence properties of the mixed strategies that result from...
Deep reinforcement learning has achieved impressive successes yet often
...
The support vector machine (SVM) is a well-established classification me...
We compare classification and regression tasks in the overparameterized
...
Agents rarely act in isolation -- their behavioral history, in particula...
We consider the stochastic linear (multi-armed) contextual bandit proble...
A continuing mystery in understanding the empirical success of deep neur...
Recent work shows unequal performance of commercial face classification
...
We introduce algorithms for online, full-information prediction that are...
Pairwise comparison data arises in many domains, including tournament
ra...