Computational Learning Theory
Learning Causal Models under Independent Changes
In many scientific applications, we observe a system in different conditions in which its components may change, rather than in isolation.
Derandomizing Multi-Distribution Learning
Multi-distribution or collaborative learning involves learning a single predictor that works well across multiple data distributions, using samples from each during training.
a8808b75b299d64a23255bc8d30fb786-Paper-Conference.pdf
Can a physicist make only a finite number of errors in the eternal quest to uncover the law of nature? This millennium-old philosophical problem, known as inductive inference, lies at the heart of epistemology.
PAC Learning Linear Thresholds from Label Proportions
The goal is to train a good instance classifier.
A Near-optimal Algorithm for Learning Margin Halfspaces with Massart Noise
In this paper, we study the same problem in the Massart noise model that we now define.
The Bayesian Stability Zoo
Algorithmic stability is a major theme in learning theory, where seminal results have firmly established its close relationship with generalization. Recent research has further highlighted the intricate interplay between stability and additional properties of interest beyond statistical generalization.
On the role of entanglement and statistics in learning
For learning Boolean concept classes, we show that the entangled and separable sample complexity are polynomially related .