arXiv AI
Aug 13
Post-Training with Policy Gradients: Optimality and the Base Model Barrier
arXiv:2603. 06957v2 Announce Type: replace-cross Abstract: We study post-training linear autoregressive models with outcome and process rewards.
The Statistical Benefits of Multiple Responses for Learning from Demonstrations
Related stories
arXiv Machine Learning
Sep 15
Bandits with Probing: Optimal Regret and the Limits of Winner Feedback
arXiv:2609.15248v1 Announce Type: new Abstract: A learner probes at most $k$ of $n$ arms each round, receives the maximum of their rewards in $[0,1]$, and competes with the best fixed arm. When does...
arXiv Machine Learning
Jun 2
Online Learning in MDPs with Partially Adversarial Transitions and Losses
arXiv:2602. 09474v2 Announce Type: replace Abstract: We study reinforcement learning in MDPs whose transition function is stochastic at most steps but may behave adversarially at a fixed subset of $\Lambda$ steps per episode.
arXiv Machine Learning
5d ago
Bandit Multiclass PAC Learning: Structural Baselines and a Confidence Direct Sum
arXiv:2609. 29694v2 Announce Type: replace Abstract: How much does high confidence cost when a multiclass learner sees only whether its prediction is correct?
arXiv Machine Learning
Aug 7
Optimal Rates for Learning with Monotone Adversaries
arXiv:2608. 06337v1 Announce Type: cross Abstract: A monotone adversary observes an i.
arXiv Machine Learning
Oct 16
Transfer Learning in Bandits from Unknown Adaptive Policies: Problem-Dependent Lower and Upper Bounds
The paper investigates transfer learning in stochastic multi‑armed bandits when historical data come from an unknown adaptive policy. It establishes asymptotic problem‑dependent lower bounds and introduces the KL‑UCB‑Transfer algorithm, which combines online and historical data to achieve logarithmic regret with an improved constant under certain mean‑gap conditions. The study also relaxes the source‑target discrepancy assumption to a one‑sided condition, demonstrating that logarithmic regret can still be achieved under misspecification and identifying regimes of improvement, degradation, or loss of guarantee.
arXiv Machine Learning
Sep 30
Optimal Multi-Reward Reinforcement Learning
arXiv:2609.36486v1 Announce Type: new Abstract: We study an unknown-transition finite-horizon Markov decision process (MDP) with a finite collection of known reward functions $\{r^1, r^2, \ldots, r^M...
arXiv AI
Jun 11
Pass@K Policy Optimization: Solving Harder Reinforcement Learning Problems
arXiv:2505. 15201v5 Announce Type: replace-cross Abstract: Reinforcement Learning (RL) algorithms sample multiple n>1 solution attempts for each problem and reward them independently.
arXiv Machine Learning
Aug 10
Multiscale Reward Hedging from Correct Demonstrations
arXiv:2608. 06825v1 Announce Type: new Abstract: Learning from correct demonstrations is harder than supervised learning when many answers are correct: after predicting, the learner sees one valid answer but not whether its own answer was valid, nor any reward.
arXiv Machine Learning
Jun 19
Matching Markets meet Cumulative Prospect Theory: Towards Optimal and Adversarially Robust Learning
arXiv:2606. 19883v1 Announce Type: new Abstract: We study a multi-agent multi-armed bandit problem in the competitive setup with two-sided matching markets under a human centric decision making model.
arXiv Machine Learning
Oct 3
Minimax Optimal Regret for Causal Logistic Bandits with Counterfactual Fairness
arXiv:2610. 01377v1 Announce Type: new Abstract: We study causal logistic bandits with counterfactual fairness constraints.
arXiv Machine Learning
Jul 1
A Complete Characterization of Learnability for Adversarial Noisy Bandits
arXiv:2605. 09200v2 Announce Type: replace Abstract: We study adversarial noisy bandits given a known function class $\mathcal{F}$.