10 papers
Toward Simultaneously Optimal Regret in U-Calibration
Rafael Frongillo, Haipeng Luo, Nishant A. Mehta +1
U-calibration studies online forecasting algorithms whose predictions can be consumed by any unknown downstream agent, guaranteeing sublinear regret simultaneously for all proper l…
Generalized Distributional Alignment Games for Unbiased Answer-Level Fine-Tuning
Mehryar Mohri, Jon Schneider, Yutao Zhong
The Distributional Alignment Game framework provides a powerful variational perspective on Answer-Level Fine-Tuning (ALFT). However, standard algorithms for these games rely on est…
Distributional Alignment Games for Answer-Level Fine-Tuning
Mehryar Mohri, Jon Schneider, Yifan Wu
We focus on the problem of \emph{Answer-Level Fine-Tuning} (ALFT), where the goal is to optimize a language model based on the correctness or properties of its final answers, rathe…
Next-Token Prediction and Regret Minimization
Mehryar Mohri, Clayton Sanford, Jon Schneider +2
We consider the question of how to employ next-token prediction algorithms in adversarial online decision-making environments. Specifically, if we train a next-token prediction mod…
Efficient Opportunistic Approachability
Teodor Vanislavov Marinov, Mehryar Mohri, Princewill Okoroafor +2
We study the problem of opportunistic approachability: a generalization of Blackwell approachability where the learner would like to obtain stronger guarantees (i.e., approach a sm…
Near-optimal Swap Regret Minimization for Convex Losses
Lunjia Hu, Jon Schneider, Yifan Wu
We give a randomized online algorithm that guarantees near-optimal expected swap regret against any sequence of adaptively chosen Lipschitz convex losse…
Swap Regret Minimization Through Response-Based Approachability
Ioannis Anagnostides, Gabriele Farina, Maxwell Fishelson +2
We consider the problem of minimizing different notions of swap regret in online optimization. These forms of regret are tightly connected to correlated equilibrium concepts in gam…
High-Dimensional Calibration from Swap Regret
Maxwell Fishelson, Noah Golowich, Mehryar Mohri +1
We study online calibration of multi-dimensional forecasts over an arbitrary convex set relative to an arbitrary norm . We connect this to externa…
Full Swap Regret and Discretized Calibration
Maxwell Fishelson, Robert Kleinberg, Princewill Okoroafor +3
We study the problem of minimizing swap regret in structured normal-form games. Players have a very large (potentially infinite) number of pure actions, but each action has an embe…
Efficient Learning and Computation of Linear Correlated Equilibrium in General Convex Games
Constantinos Daskalakis, Gabriele Farina, Maxwell Fishelson +2
We propose efficient no-regret learning dynamics and ellipsoid-based methods for computing linear correlated equilibria$\unicode{x2014}$a relaxation of correlated equilibria and a…