The paper introduces Deep-BQRL, a model‑free distributional reinforcement‑learning framework that extends buffered‑quantile learning to neural function approximation. It learns conditional return quantiles from sampled transitions, constructs buffered action scores, and uses ensemble disagreement for exploration, enabling risk‑sensitive decision‑making without explicit return‑law planning. Experiments on asset‑selling and slippery FrozenLake show that Deep‑BQRL achieves smaller mean cumulative point‑quantile policy gaps than PPO and TRPO, while illustrating interpretable risk‑sensitive stopping decisions.
By Mohammad Alipour-vaezi, Sajad Khodadadian
arXiv:2611.05847v1 Announce Type: cross
Abstract: Joint-Embedding Predictive Architectures (JEPAs) support planning by comparing predicted latent states with an encoded goal. For this comparison to g...
By Ruilong Yu, Fei Ye
arXiv:2509. 10303v2 Announce Type: replace-cross Abstract: Online reinforcement learning (RL) approaches have demonstrated strong performance on Job Shop Scheduling (JSP) and Flexible JSP (FJSP) problems by learning scheduling policies through direct interaction with simulated environments.
By Jesse van Remmerden, Zaharah Bukhsh, Yingqian Zhang
arXiv:2607. 26924v1 Announce Type: new Abstract: Recent work on LeWorldModel (LeWM) has shown that the Sketched Isotropic Gaussian Regularizer (SIGReg) enables stable end-to-end world-model learning from pixels by regularizing the latent marginal distribution toward an isotropic Gaussian, thereby preventing representation collapse.
By Chang Liu, Fei Suo, Yanzhou Jin, Yusuke Iwasawa, Yutaka Matsuo, Yaonan Zhu
Recent work on LeWorldModel (LeWM) has shown that the Sketched Isotropic Gaussian Regularizer (SIGReg) enables stable end-to-end world-model learning from pixels by regularizing the latent marginal distribution toward an isotropic Gaussian, thereby preventing representation collapse. While effective and elegant in single-task settings, this recipe does not extend reliably to multi-task training, leading to substantially worse downstream behavior-cloning performance.
The paper introduces the concept of decision‑metric alignment, which ensures that Euclidean distance to a goal latent in JEPA‑style latent world models correctly ranks action sequences for model‑predictive control. It proposes two metrics—Plan‑Real Spearman and CEM‑stage Spearman—to evaluate latent–real rank agreement, and identifies encoder distortion, terminal rollout error, and candidate margins as key factors affecting alignment. Building on these insights, the authors present DA‑LeWM, an enhanced latent world model that incorporates inverse‑dynamics and demonstration‑conditioned goal‑action heads, leading to faster convergence and higher online success rates compared to the baseline LeWM while maintaining similar probe scores.
By Jiawei Wang, Ke Rui, Yushen Zuo, Yichun Feng, Minglei Li
The paper introduces Q-Target Pretrained Transformers (QTPT), a method that replaces supervised behavior cloning with a Bellman-style Q‑target objective for in‑context reinforcement learning. QTPT retains the context‑conditioned Transformer architecture but learns to estimate action values using rewards and transitions from the context, rather than merely imitating offline actions. The authors provide theoretical analysis in stochastic linear bandits and finite‑horizon MDPs, demonstrating improved robustness to weak or suboptimal data, and empirically show gains over supervised pretraining on controlled RL benchmarks and extensions to D4RL Kitchen and AntMaze.
By Yichen Lin, Xuyuan Xiong, Xue Wang, Xiangfu Meng, Mike Mingcheng Wei, Tao Yao
arXiv:2608. 12973v3 Announce Type: replace-cross Abstract: In this paper, we study online statistical inference for quantile-based distributional policy evaluation, which aims to characterize the distribution of discounted cumulative rewards under a given policy.
By Zijie Cheng, Yang Peng, Zhihua Zhang
arXiv:2607. 28916v1 Announce Type: cross Abstract: Multistep credit assignment is critical for sample-efficient reinforcement learning, yet managing off-policy bias in Q-learning remains a fundamental challenge.
By Brett Daley
arXiv:2610.27280v1 Announce Type: cross
Abstract: Offline goal-conditioned reinforcement learning enables skill acquisition by reaching any goal from any state without dense rewards. Model-based meth...
By Yunzhe Hu, Dong Xu
arXiv:2608. 00978v1 Announce Type: new Abstract: Flow Matching trains continuous-time generative models by regressing the velocity field of a probability path between a simple source distribution and a target data distribution.
By Jin-Young Kim, So-Yoon Cho, Hyun-Gyoon Kim
The paper investigates how latent world models (specifically JEPA-style models) use Euclidean distance to a goal latent as a cost for model‑predictive control (MPC). It introduces two metrics—Plan‑Real Spearman and CEM‑stage Spearman—to evaluate how well latent‑space distances align with real‑task progress, a property termed decision‑metric alignment. By identifying encoder distortion, terminal rollout error, and candidate margins as key factors, the authors propose DA‑LeWM, which augments the base model with inverse‑dynamics and demonstration‑conditioned goal‑action heads, leading to faster convergence and higher online success while maintaining similar probe scores.