Adaptive Correlation-Weighted Intrinsic Rewards for Reinforcement Learning

2026-02-27 · Viet Bac Nguyen, Phuong Thai Nguyen arxiv

We propose ACWI (Adaptive Correlation Weighted Intrinsic), an adaptive intrinsic reward scaling framework designed to dynamically balance intrinsic and extrinsic rewards for improved exploration in sparse reward reinforcement learning. Unlike conventional approaches that rely on manually tuned scalar coefficients, which often result in unstable or suboptimal performance across tasks, ACWI learns a state dependent scaling coefficient online. Specifically, ACWI introduces a lightweight Beta Network that predicts the intrinsic reward weight directly from the agent state through an encoder based architecture. The scaling mechanism is optimized using a correlation based objective that encourages alignment between the weighted intrinsic rewards and discounted future extrinsic returns. This formulation enables task adaptive exploration incentives while preserving computational efficiency and training stability. We evaluate ACWI on a suite of sparse reward environments in MiniGrid. Experimental results demonstrate that ACWI consistently improves sample efficiency and learning stability compared to fixed intrinsic reward baselines, achieving superior performance with minimal computational overhead.

📄 PDF Abstract BibTeX arXiv:2602.24081

Code (0)

등록된 구현이 없습니다.

Tasks

Computational Efficiency Reinforcement Learning

Deep Learning of Intrinsically Motivated Options in the Arcade Learning Environment

2021-09-29 · Louis Bagot, Kevin Mets, Tom De Schepper, Peter Hellinckx 외

Although Intrinsic Motivation allows a Reinforcement Learning agent to generate directed behaviors in an environment, even with sparse or noisy rewards, combining intrinsic and extrinsic rewards is non trivial. As an alt…

Atari Games Benchmarking Deep Reinforcement Learning Fairness+2

Learning Intrinsically Motivated Options to Stimulate Policy Exploration

2020-06-12 · ICML Workshop LifelongML 2020 7 · Louis Bagot, Kevin Mets, Steven Latré

A Reinforcement Learning (RL) agent needs to find an optimal sequence of actions in order to maximize rewards. This requires consistent exploration of states and action sequences to ensure the policy found is optimal. On…

reinforcement-learning Reinforcement Learning (RL)

A Temporally Correlated Latent Exploration for Reinforcement Learning

2024-12-06 · SuMin Oh, Wansoo Kim, HyunJin Kim

Efficient exploration remains one of the longstanding problems of deep reinforcement learning. Instead of depending solely on extrinsic rewards from the environments, existing methods use intrinsic rewards to enhance exp…

Deep Reinforcement Learning Efficient Exploration reinforcement-learning Reinforcement Learning

Mol-AIR: Molecular Reinforcement Learning with Adaptive Intrinsic Rewards for Goal-directed Molecular Generation

2024-03-29 · Jinyeong Park, Jaegyoon Ahn, Jonghwan Choi, Jibum Kim

Optimizing techniques for discovering molecular structures with desired properties is crucial in artificial intelligence(AI)-based drug discovery. Combining deep generative models with reinforcement learning has emerged …

Drug Discovery reinforcement-learning Reinforcement Learning

From Kicking to Causality: Simulating Infant Agency Detection with a Robust Intrinsic Reward

2025-07-20 · Xia Xu, Jochen Triesch arxiv

While human infants robustly discover their own causal efficacy, standard reinforcement learning agents remain brittle, as their reliance on correlation-based rewards fails in noisy, ecologically valid scenarios. To addr…

Reinforcement Learning Causal Inference