arXiv research

New algorithms for generalized linear bandits with unknown reward functions.

New algorithms improve best-arm identification with varying rewards.

CoCoRL learns safe constraints from demonstrations with unknown rewards.

Consider a nonparametric contextual multi-arm bandit problem where each arm $a \in [K]$ is associated to a nonparametric reward function $f_a: [0,1] \to \mathbb{R}$ mapping from contexts to the expected reward. Suppose that there is a large set of arms, yet there is a simple but unknown structure amongst the arm reward…

Algorithm learns user's reward function from hypothetical behaviors.

The paper tackles resource allocation for arms with unknown and random rewards, achieving optimal regret bounds.

Reinforcement learning with trajectory feedback instead of state-action rewards.

Study resource allocation strategies in sequential decisions with unknown rewards.

Study reward-free RL in non-linear settings, improving efficiency and removing assumptions.

Study noisy rewards in online decision-making with unknown distributions.

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

New RL algorithm tackles adversarial RMAB with unknown transitions and bandit feedback.

We consider a class of restless multi-armed bandit (RMAB) problems with unknown arm dynamics. At each time, a player chooses an arm out of N arms to play, referred to as an active arm, and receives a random reward from a finite set of reward states. The reward state of the active arm transits according to an unknown Ma…

We here adopt Bayesian nonparametric mixture models to extend multi-armed bandits in general, and Thompson sampling in particular, to scenarios where there is reward model uncertainty. In the stochastic multi-armed bandit, the reward for the played arm is generated from an unknown distribution. Reward uncertainty, i.e.…

New algorithm learns safe policies in unknown environments.

Paper tackles unknown variances in best-arm identification.

In structured output prediction tasks, labeling ground-truth training output is often expensive. However, for many tasks, even when the true output is unknown, we can evaluate predictions using a scalar reward function, which may be easily assembled from human knowledge or non-differentiable pipelines. But searching th…

Optimal strategy proposed for maximizing cumulative reward in continuum-armed bandits.

Reinforcement learning is concerned with identifying reward-maximizing behaviour policies in environments that are initially unknown. State-of-the-art reinforcement learning approaches, such as deep Q-networks, are model-free and learn to act effectively across a wide range of environments such as Atari games, but requ…

New algorithm reduces dynamic regret for MDPs with unknown transition and adversarial rewards.

Paper proposes a method to learn and exceed expert demonstrations in unknown reward environments.

Optimistic Thompson Sampling reduces regret in unknown multi-player games.

We tackle the problem of acting in an unknown finite and discrete Markov Decision Process (MDP) for which the expected shortest path from any state to any other state is bounded by a finite number $D$. An MDP consists of $S$ states and $A$ possible actions per state. Upon choosing an action $a_t$ at state $s_t$, one re…

GACBO optimizes unknown causal graphs with interventions.

New RL approach learns dynamic VCG mechanisms in unknown MDP environments.

New strategy optimally identifies best arm in unknown variance Gaussian bandits.

Inverse Reinforcement Learning (IRL) describes the problem of learning an unknown reward function of a Markov Decision Process (MDP) from observed behavior of an agent. Since the agent's behavior originates in its policy and MDP policies depend on both the stochastic system dynamics as well as the reward function, the …

This paper addresses reward estimation and incentive design for agents with hidden rewards.

Thompson sampling used for linear bandits with normal-gamma priors.

We extend Bayesian multi-armed bandit (MAB) algorithms beyond their original setting by making use of sequential Monte Carlo (SMC) methods. A MAB is a sequential decision making problem where the goal is to learn a policy that maximizes long term payoff, where only the reward of the executed action is observed. In the …

Algorithm aggregates rewards from multiple players to learn related tasks in online bandit learning.

PQR estimates reward functions from actions and states without assuming state-only rewards.

B-REX efficiently learns Atari game policies from pixel inputs using Bayesian methods.

This paper improves understanding of GAIL's generalization and computational efficiency.

Learning to make decisions from observed data in dynamic environments remains a problem of fundamental importance in a number of fields, from artificial intelligence and robotics, to medicine and finance. This paper concerns the problem of learning control policies for unknown linear dynamical systems so as to maximize…

New method for linear bandits with unknown sparsity, improving sparse regret bounds.

In a linear stochastic bandit model, each arm is a vector in an Euclidean space and the observed return at each time step is an unknown linear function of the chosen arm at that time step. In this paper, we investigate the problem of learning the best arm in a linear stochastic bandit model, where each arm's expected r…

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 …

Study on reward poisoning attacks on CMAB, revealing attackability depends on adversary's knowledge.

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

The paper develops a robust algorithm for contextual bandits with heavy-tailed rewards.

This paper proposes DeepSynth, a method for effective training of deep Reinforcement Learning (RL) agents when the reward is sparse and non-Markovian, but at the same time progress towards the reward requires achieving an unknown sequence of high-level objectives. Our method employs a novel algorithm for synthesis of c…

We study the Combinatorial Pure Exploration problem with Continuous and Separable reward functions (CPE-CS) in the stochastic multi-armed bandit setting. In a CPE-CS instance, we are given several stochastic arms with unknown distributions, as well as a collection of possible decisions. Each decision has a reward accor…

New algorithm reduces regret for single-index bandits to nearly optimal.

We consider the problem of learning to play a repeated multi-agent game with an unknown reward function. Single player online learning algorithms attain strong regret bounds when provided with full information feedback, which unfortunately is unavailable in many real-world scenarios. Bandit feedback alone, i.e., observ…

Reinforcement learning agents are prone to undesired behaviors due to reward mis-specification. Finding a set of reward functions to properly guide agent behaviors is particularly challenging in multi-agent scenarios. Inverse reinforcement learning provides a framework to automatically acquire suitable reward functions…

Improved regret for single-index bandits with optimal algorithm.

We consider a problem of learning the reward and policy from expert examples under unknown dynamics. Our proposed method builds on the framework of generative adversarial networks and introduces the empowerment-regularized maximum-entropy inverse reinforcement learning to learn near-optimal rewards and policies. Empowe…