arXiv AI
Jun 16
Safe Exploration via Policy Priors
arXiv:2601. 19612v3 Announce Type: replace-cross Abstract: Safe exploration is a key requirement for reinforcement learning (RL) agents to learn and adapt online, beyond controlled (e.
Safe Learning Under Irreversible Dynamics via Asking for Help
The paper presents an algorithm that lets a learning agent ask for help from a mentor and transfer knowledge between similar states, enabling safe and effective learning in Markov decision processes with irreversible dynamics and infinite state spaces. It proves that both regret and the number of mentor queries grow sublinearly over time, using a sequence of three reductions to achieve a general result. The work claims to be the first formal proof that an agent can achieve high reward while becoming self‑sufficient in an unknown, unbounded, high‑stakes environment without resets.
Related stories
arXiv Machine Learning
Aug 3
Beyond Black-Box Advice: Learning-Augmented Algorithms for MDPs with Q-Value Predictions
arXiv:2307. 10524v3 Announce Type: replace Abstract: We study the tradeoff between consistency and robustness in the context of a single-trajectory time-varying Markov Decision Process (MDP) with untrusted machine-learned advice.
arXiv AI
Jul 10
Provably Optimal Learning Algorithms for Assistance Games
arXiv:2607. 08012v1 Announce Type: cross Abstract: This paper studies an online variant of the assistance games framework, where an informed agent and an uninformed agent repeatedly interact over $T$ timesteps to optimize a common reward function.
arXiv AI
Sep 2
Independent Reinforcement Learning in Discounted Markov Games
arXiv:2609.00504v1 Announce Type: cross Abstract: In this work, we study radically uncoupled learning in discounted general-sum Markov games. Assuming ``$\mathsf{ETH}$ for $\mathsf{PPAD}$", we show t...
arXiv Machine Learning
Aug 11
Regret, equilibrium, and learning in games: A guided tour
arXiv:2608. 09389v1 Announce Type: cross Abstract: This note aims to serve as an entry point to the literature on learning in games, a topic with significant theoretical appeal and a wide range of applications -- from machine learning and data science to economics and beyond.
arXiv AI
Aug 21
Adaptive Probabilistic Shielding by Learning MDPs for Safe Reinforcement Learning
arXiv:2608. 19836v1 Announce Type: cross Abstract: Probabilistic shielding is a technique for safe reinforcement learning (RL).
arXiv Machine Learning
Sep 10
A First-Order Learning Algorithm for Online Resource Allocation with Constant Regret
arXiv:2609.05895v1 Announce Type: new Abstract: We study a finite-horizon online resource allocation problem with initial resource capacities proportional to the horizon. In each period, a request ty...
arXiv Machine Learning
Oct 7
On the Computational Tractability of Robust Bandits
The paper studies robust bandits, a framework for learning when the true environment lies outside the learner’s hypothesis class. It presents a polynomial‑time algorithm achieving ≠O(√T) regret for a specific tractable case, while showing that slight extensions of this case are NP‑hard. The work highlights the computational limits of robust bandits and suggests their relevance to AI alignment.
arXiv Machine Learning
Jul 21
Theoretical Foundations of $\max$@$k$ Reinforcement Learning
arXiv:2607. 17823v1 Announce Type: new Abstract: Reinforcement Learning is a cornerstone technique for modern large reasoning models.
arXiv AI
Jun 2
Robust Shielding for Safe Reinforcement Learning
arXiv:2606. 00270v1 Announce Type: new Abstract: Shielding is an effective approach to formally guarantee the safety of reinforcement learning agents in Markov decision processes (MDPs).
arXiv Statistics ML
6d ago
Constrained Gradient Equilibrium: Approachability under Locally Constrained Adversaries
arXiv:2610.33985v1 Announce Type: cross Abstract: Gradient equilibrium (GEQ) is a recent framework in online learning that requires the time-averaged gradients along the learner's trajectory to vanis...
arXiv Machine Learning
Oct 21
Confessions of a Masked MDP
arXiv:2610.28360v1 Announce Type: cross Abstract: In reinforcement learning, one presumes the action set to be known ahead of time, and, thus, one must know if any state-action pairs are not permissi...