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
The paper introduces state abstractions that preserve the difference of Q‑functions for offline reinforcement learning, aiming to exclude irrelevant dynamics from rich state data. It proposes a dynamic generalization of the R‑learner that uses orthogonal estimation and sparse learning to estimate the Q‑function contrast, achieving faster convergence and consistency under a margin condition. Experiments on simulated and simulator‑augmented real data show variance reductions and demonstrate that the necessary information for sequential decision‑making can be smaller than that required for full state prediction.
By Defu Cao, Angela Zhou
Pre-training followed by fine-tuning has become the dominant recipe for learning performant policies, and in value-based reinforcement learning (RL) this raises a natural question: given a pretrained policy, should the Q-function be pretrained on offline data too? Conventional wisdom suggests it should, but recent results show that online RL with a randomly-initialized Q-function can result in highly performant and reliable policies without needing to pretrain the Q-function.
The paper introduces BUMEX, a reinforcement learning exploration strategy that leverages a set of prior models containing the true transition kernel and reward function. By optimizing over this model set, the method derives upper and lower bounds on the Q‑function to guide exploration, providing theoretical guarantees of convergence to the optimal policy. When the model set follows a bounded‑parameter MDP structure, the optimization becomes convex, enabling finite‑time convergence under mild assumptions and demonstrating accelerated learning in simulations.
By J. S. van Hulst, W. P. M. H. Heemels, D. J. Antunes
In value-based reinforcement learning, improving the accuracy of policy evaluation has been shown to improve downstream policy optimization performance. The widely adopted family of approximations rel...
arXiv:2608. 02034v1 Announce Type: new Abstract: Multi-step returns accelerate reward propagation in off-policy reinforcement learning, but couple the evaluation of each decision to the suboptimal logged actions that follow it, inducing a pessimistic bias that grows with the horizon.
By Abdelghani Ghanem, Mounir Ghogho
arXiv:2608. 09805v1 Announce Type: cross Abstract: Exploration has been a focus of reinforcement learning research for a long time.
By Vatsal Venkatkrishna, Nico Daheim, Iryna Gurevych
arXiv:2607. 27203v1 Announce Type: new Abstract: Pre-training followed by fine-tuning has become the dominant recipe for learning performant policies, and in value-based reinforcement learning (RL) this raises a natural question: given a pretrained policy, should the Q-function be pretrained on offline data too?
By Perry Dong, Ron Polonsky, Dorsa Sadigh, Chelsea Fin
The paper introduces Reward Stimulation Implicit Q-Learning (RSIQL), a non-hierarchical approach to improve offline goal-conditioned reinforcement learning. RSIQL adds auxiliary reward signals at intermediate states that are predicted to aid progress toward the goal, thereby reducing the delay in training supervision. Experiments on D4RL goal-reaching benchmarks and OGBench demonstrate that RSIQL outperforms baseline goal-conditioned IQL and rivals hierarchical offline methods while maintaining a simple flat policy structure.
By Jing Zhang
arXiv:2609.36178v1 Announce Type: cross
Abstract: Group Relative Policy Optimization (GRPO) has become a promising approach for training large language model agents. However, its uniform assignment o...
By Dongwon Jung, Hemanth Neelgund Ramesh, Yifan Wang, Xiaomin Li, Yuexing Hao, Yu Hu, Muhao Chen, Varun Chandrasekaran, Andrzej Banburski-Fahey, Jaron Lanier
arXiv:2605. 16103v2 Announce Type: replace Abstract: Q-learning is known to suffer from overestimation bias: because the Bellman update maximizes noisy or imperfect action-value estimates, positive errors can be selected and propagated, causing learned values to exceed the true optimal values.
By Donghwan Lee
arXiv:2602. 05459v2 Announce Type: replace Abstract: Offline goal-conditioned reinforcement learning (GCRL) is typically benchmarked by the best tuned success rate of each method.
By Jan Malte T\"opperwien, Aditya Mohan, Marius Lindauer