跳到正文
arXiv cs.AI· Jingyang You, Hanna Kurniawati·· 10 小时前AI 评分31

GLiBRL:用广义线性任务模型实现精确贝叶斯推理的元强化学习框架

Meta-RL with Bayesian Linear Task Models

AI 导读

GLiBRL 是一个深度贝叶斯强化学习框架,将广义线性任务模型与可学习非线性基函数结合,通过共轭贝叶斯推理实现任务参数与模型噪声的精确序列后验更新,并用闭式边际似然替代变分推断。

正文 · 原文

View PDF HTML (experimental)

Abstract:Deep Bayesian reinforcement learning adapts to unseen tasks by inferring latent transition and reward models, but existing methods typically rely on variational posteriors and evidence lower bounds, introducing approximation error and unstable task representations. We introduce GLiBRL, a deep Bayesian RL framework that combines generalised linear task models with learnable non-linear basis functions. GLiBRL features conjugate Bayesian inference, yielding exact, sequential posterior updates over task parameters and model noise, together with a closed-form marginal likelihood that eliminates variational inference. The update is naturally permutation-invariant, allowing GLiBRL to integrate with both off- and on-policy algorithms. GLiBRL also learns task representation admitting an exact kernel identity, relating distances between task representations to kernel discrepancies over the task contexts. Compared against eight representative or recent meta reinforcement learning methods, GLiBRL achieves the highest aggregate zero-shot test performance on both the MuJoCo locomotion and MetaWorld manipulation benchmarks.
Comments: Accepted at NeurIPS 2026
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Robotics (cs.RO)
Cite as: arXiv:2512.20974 [cs.LG]
  (or arXiv:2512.20974v5 [cs.LG] for this version)
  https://doi.org/10.48550/arXiv.2512.20974

arXiv-issued DOI via DataCite

Submission history

From: Jingyang You [view email]
[v1] Wed, 24 Dec 2025 06:00:51 UTC (3,218 KB)
[v2] Thu, 7 May 2026 05:25:01 UTC (4,837 KB)
[v3] Fri, 8 May 2026 01:20:50 UTC (4,834 KB)
[v4] Wed, 9 Sep 2026 09:27:59 UTC (1,420 KB)
[v5] Tue, 17 Nov 2026 23:10:19 UTC (1,299 KB)

来源:arXiv cs.AI · arxiv.org