Peter Dixon — accepted AI-conference papers

2024

Replicability in Learning: Geometric Partitions and KKM-Sperner Lemma

Jason Vander Woude, Peter Dixon, A. Pavan, Jamie Radcliffe, N. V. Vinodchandran

NeurIPS 2024poster

This paper studies replicability in machine learning tasks from a geometric viewpoint. Recent works have revealed the role of geometric partitions and Sperner's lemma (and its variations) in designing replicable learning algorithms and in establishing impossibility results. A partition $\mathcal…

Cited by 0 Source ↗PDF ↗Scholar ↗

2023

List and Certificate Complexities in Replicable Learning

Peter Dixon, A. Pavan, Jason Vander Woude, N V Vinodchandran

NeurIPS 2023spotlight

We investigate replicable learning algorithms. Informally a learning algorithm is replicable if the algorithm outputs the same canonical hypothesis over multiple runs with high probability, even when different runs observe a different set of samples from the unknown data distribution. In general, s…

Cited by 14 Source ↗PDF ↗Scholar ↗