← Back to all news
arXiv AI By Jingyang You, Hanna Kurniawati

Meta-RL with Bayesian Linear Task Models

Read the original on arXiv AI →

The paper introduces GLiBRL, a deep Bayesian reinforcement learning framework that uses generalized linear task models with learnable nonlinear basis functions. GLiBRL performs exact, sequential Bayesian inference without variational approximations, providing closed‑form posterior updates and a marginal likelihood. The method is permutation‑invariant, compatible with both off‑ and on‑policy algorithms, and achieves the best zero‑shot performance among eight meta‑RL baselines on MuJoCo locomotion and MetaWorld manipulation tasks.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

Related stories

arXiv Machine Learning
Oct 15

Active Exploration in Bayesian Model-based Reinforcement Learning for Robot Manipulation

The paper presents a Bayesian deep learning approach for active exploration in model-based reinforcement learning applied to robot manipulation. By improving model calibration and actively selecting transitions that maximize information gain, the method enhances data efficiency during a preliminary exploration phase. Experiments on a realistic robotic arm simulator demonstrate that the approach achieves comparable performance to existing methods while requiring significantly fewer robot executions.

By Carlos Plou, Ana C. Murillo, Ruben Martinez-Cantin
reinforcement-learningrobotics
More like this →
arXiv AI
Oct 26

PrefO2O: Zero-Shot Preference-Based Reinforcement Learning via Offline-to-Online Latent Preference Transfer

PrefO2O introduces an offline-to-online latent preference transfer framework that enables zero‑shot preference‑based reinforcement learning on unseen tasks without human feedback. It learns a task‑invariant latent preference space from limited offline preference data, disentangling transferable preference structures from task‑specific variations. The method uses confidence‑weighted optimal transport to generate reliable pseudo‑preference supervision for online reward adaptation, achieving performance that matches or exceeds traditional PbRL methods trained with real human feedback across various robotic manipulation and locomotion benchmarks.

By Ruiqi Wang, Dezhong Zhao, Xuesu Xiao, Erdem Biyik, Byung-Cheol Min, Guohua Chen
reinforcement-learningroboticsbenchmarkssafety
More like this →