arXiv research

This work analyzes the value of future reward information in RL.

New complexity measure MEHC refines MDP upper bounds and rewards informativeness.

We propose a generic, Bayesian, information geometric approach to the exploration--exploitation trade-off in multi-armed bandit problems. Our approach, BelMan, uniformly supports pure exploration, exploration--exploitation, and two-phase bandit problems. The knowledge on bandit arms and their reward distributions is su…

Study shows mutual information can reward structure learning agents without expert systems.

ARL uses queries to learn rewards, focusing on cost vs. reward value.

Thompson Sampling bounds for contextual bandits with sub-Gaussian rewards.

Paper simplifies complex AI exploration by predicting future rewards.

New RL algorithms use lookahead info to maximize rewards.

Designers of AI agents often iterate on the reward function in a trial-and-error process until they get the desired behavior, but this only guarantees good behavior in the training environment. We propose structuring this process as a series of queries asking the user to compare between different reward functions. Thus…

The paper analyzes the reward improvement of aligned policies in large language models.

The paper proposes a model reward scheme for collaborative ML based on Shapley value and information gain.

Reinforcement learning for embodied agents is a challenging problem. The accumulated reward to be optimized is often a very rugged function, and gradient methods are impaired by many local optimizers. We demonstrate, in an experimental setting, that incorporating an intrinsic reward can smoothen the optimization landsc…

Algorithm learns user's reward function from hypothetical behaviors.

The paper enhances preference learning by incorporating response time data.

Paper tackles robustness in reward learning with partial identifiability.

New model for bandit problem with linear rewards and side information.

This paper extends the MAB problem to consider risk-reward tradeoffs.

We consider an adversarial online learning setting where a decision maker can choose an action in every stage of the game. In addition to observing the reward of the chosen action, the decision maker gets side observations on the reward he would have obtained had he chosen some of the other actions. The observation str…

Proper balance between exploitation and exploration is what makes good decisions, which achieve high rewards like payoff or evolutionary fitness. The Infomax principle postulates that maximization of information directs the function of diverse systems, from living systems to artificial neural networks. While specific a…

Paper designs a bandit algorithm without reward distribution info.

The paper introduces a health-informed policy gradient method for multi-agent reinforcement learning.

Proposes learning latent reward model for planning from rewards.

The paper tackles multi-player information asymmetry bandits in metric spaces.

Algorithm reduces regret in multi-player bandits with unknown collision rewards.

Unified approach combines reward maximization and empowerment for RL.

In this paper we propose a novel framework for decentralized, online learning by many learners. At each moment of time, an instance characterized by a certain context may arrive to each learner; based on the context, the learner can select one of its own actions (which gives a reward and provides information) or reques…

New method for contextual bandit with missing rewards.

A new method resolves non-identifiability in reward modeling using anchor labels.

Algorithm optimizes multi-player learning with noisy rewards without direct communication.

In this paper, we propose an information-theoretic exploration strategy for stochastic, discrete multi-armed bandits that achieves optimal regret. Our strategy is based on the value of information criterion. This criterion measures the trade-off between policy information and obtainable rewards. High amounts of policy …

Enhances reward specification in RL with a novel language-based approach.

New algorithm DUSA minimizes regret in structured bandits with structural information.

TRAIL improves robot imitation learning by focusing on task-relevant features.

New algorithm optimizes for long-term user satisfaction in delayed reward settings.

BerlinUCB learns from episodic rewards in nonstationary contexts.

Study on stochastic one-sided full-information bandit problem.

Reward collapse occurs when ranking-based reward models yield uniform rewards for different prompts.

This work improves wireless network learning by using side-information about interference.

New algorithm achieves optimal regret in average reward MDPs without prior bias information.

Optimizing rewards under budget constraints with correlated costs and rewards.

BatchGFN uses generative flow networks for efficient batch active learning.

This study proposes hidden state curiosity to enhance RL models' resilience against noise.

Estimates rewards from a demonstrator's learning process.

The paper explores how regularization can lead to convergence in imperfect information games.

A new learning method for prosthetic arms without explicit rewards.

IDS optimizes regret in stochastic partial monitoring with linear rewards.

Adaptive MAB algorithms handle composite, anonymous feedback without reward interval knowledge.

MERL uses evolutionary and gradient-based methods to optimize sparse team-based and dense agent-specific rewards in multiagent coordination.