4 papers · 1 filter
Post-Training at the Edge of Detectability: A Game-Theoretic Approach to Fine-Tuning
Keegan Harris, Brian W. Lee, Ian Waudby-Smith +3
Reinforcement learning (RL) fine-tuning is widely used in language model training to improve performance on a target task while limiting drift from a reference policy. A standard w…
Blackwell Approachability and Gradient Equilibrium are Equivalent
Brian W. Lee, Nika Haghtalab, Michael I. Jordan +1
Gradient equilibrium (GEQ) is a recently introduced online optimization framework that generalizes first-order stationarity from offline optimization and abstracts problems like on…
Panprediction: Optimal Predictions for Any Downstream Task and Loss
Sivaraman Balakrishnan, Nika Haghtalab, Daniel Hsu +2
Supervised learning is classically formulated as training a model to minimize a fixed loss function over a fixed distribution, or task. However, an emerging paradigm instead views…
Single Trajectory Conformal Prediction
Brian Lee, Nikolai Matni
We study the performance of risk-controlling prediction sets (RCPS), an empirical risk minimization-based formulation of conformal prediction, with a single trajectory of temporall…