arXiv research

New bandit algorithm detects and adapts to seasonal changes in rewards.

New algorithm handles MDPs with unknown, changing rewards efficiently.

The multi-armed bandit (MAB) problem is a classic example of the exploration-exploitation dilemma. It is concerned with maximising the total rewards for a gambler by sequentially pulling an arm from a multi-armed slot machine where each arm is associated with a reward distribution. In static MABs, the reward distributi…

Study adapts combinatorial semi-bandit for piecewise stationary, causally related rewards.

New algorithm tracks changes in infinite action space rewards.

New method reduces dynamic regret for non-stationary bandits.

New algorithms detect changes in non-stationary MABs for better performance.

Study identifies change points in piecewise constant reward functions with fixed exploration budget.

New algorithm reduces online learning regret by exploiting historical invariances.

New method tracks significant shifts in nonparametric bandits.

DS-TS adapts to abrupt and smooth changes in bandit problems.

Multi-armed bandit (MAB) is a class of online learning problems where a learning agent aims to maximize its expected cumulative reward while repeatedly selecting to pull arms with unknown reward distributions. We consider a scenario where the reward distributions may change in a piecewise-stationary fashion at unknown …

This paper tackles RL in non-stationary environments, improving decision-making.

The multi-armed bandit problem has been extensively studied under the stationary assumption. However in reality, this assumption often does not hold because the distributions of rewards themselves may change over time. In this paper, we propose a change-detection (CD) based framework for multi-armed bandit problems und…

Proposes a new TS algorithm for non-stationary bandits using KS tests.

Algorithm improves multi-armed bandit performance by transferring reward samples.

Efficiently adapting to new environments and changes in dynamics is critical for agents to successfully operate in the real world. Reinforcement learning (RL) based approaches typically rely on external reward feedback for adaptation. However, in many scenarios this reward signal might not be readily available for the …

Self-supervised policy adapts after deployment without rewards.

Study incentivizes exploration in non-stationary MAB with compensation.

Reward tweaking optimizes behavior for long-term goals by adjusting the reward function.

We consider reinforcement learning in changing Markov Decision Processes where both the state-transition probabilities and the reward functions may vary over time. For this problem setting, we propose an algorithm using a sliding window approach and provide performance guarantees for the regret evaluated against the op…

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

Algorithm improves transfer learning by inferring successor maps.

AIRL learns robust, generalizable reward functions from demonstrations.

The paper examines how updates to probabilistic models influence behavior based on evidence.

Multi-armed bandit algorithms have become a reference solution for handling the explore/exploit dilemma in recommender systems, and many other important real-world problems, such as display advertisement. However, such algorithms usually assume a stationary reward distribution, which hardly holds in practice as users' …

We introduce a new online learning framework where, at each trial, the learner is required to select a subset of actions from a given known action set. Each action is associated with an energy value, a reward and a cost. The sum of the energies of the actions selected cannot exceed a given energy budget. The goal is to…

A new algorithm for personalized recommendations adapts to changing user interests.

New strategy for gambling with changing environments achieves sub-linear regret.

Learning robust value functions given raw observations and rewards is now possible with model-free and model-based deep reinforcement learning algorithms. There is a third alternative, called Successor Representations (SR), which decomposes the value function into two components -- a reward predictor and a successor ma…

Study scaling laws of reward model overoptimization in reinforcement learning.

Transforming sparse outcomes into dense process rewards for efficient reinforcement learning.

Current reinforcement learning (RL) methods can successfully learn single tasks but often generalize poorly to modest perturbations in task domain or training procedure. In this work, we present a decoupled learning strategy for RL that creates a shared representation space where knowledge can be robustly transferred. …

A new method shapes reinforcement learning environments by abstracting large state spaces.

New TS algorithms improve performance in non-stationary multi-armed bandit problems.

The paper introduces a method for multi-agent reinforcement learning to coordinate exploration.

LNUCB-TA improves MAB performance by dynamically adjusting exploration rates and recognizing spatiotemporal patterns.

We propose a useful approach for investigating the statistical properties of foreign currency exchange rates. Our approach is based on queueing theory, particularly, the so-called renewal-reward theorem. For the first passage processes of the Sony Bank US dollar/Japanese yen (USD/JPY) exchange rate, we evaluate the ave…

New algorithm adapts to changing action sets in sequential decision making.

New method tracks significant arm switches to improve bandit algorithms.

New algorithm detects changes in combinatorial semi-bandit rewards.

New method learns adaptive exploration strategies for dynamic tasks.

New RL algorithm tackles non-stationary environments with flexible policy updates.

We consider a multi-armed bandit problem in a setting where each arm produces a noisy reward realization which depends on an observable random covariate. As opposed to the traditional static multi-armed bandit problem, this setting allows for dynamically changing rewards that better describe applications where side inf…

Language models perform worse with implicit reward models than explicit ones.

The study analyzes how neural reward models learn features for policy optimization in a Gaussian single-index model.

In a multi-armed bandit (MAB) problem a gambler needs to choose at each round of play one of K arms, each characterized by an unknown reward distribution. Reward realizations are only observed when an arm is selected, and the gambler's objective is to maximize his cumulative expected earnings over some given horizon of…

HOFLON automates process start-ups and grade-changes using offline RL and online optimization.