Xing Lei

Xing Lei, Jincheng Wang, Xuetao Zhang et al.

Offline goal-conditioned RL (GCRL) learns goal-reaching policies from static datasets, but real-world datasets are often partially observable and history-dependent, exhibiting a mix of Markovian and non-Markovian that violate standard RL assumptions. History-aware sequence models such as Decision Transformer (DT) are a natural fit for long-term dependency modeling, yet pure attention is inefficient and brittle when handling local Markovian structure and long-range context simultaneously. Although recent hybrid architectures (e.g., LSDT) introduce local extractors to improve local dependencies modeling, the fixed-window extraction cannot adapt its effective memory to varying dependency lengths in temporally heterogeneous settings, often truncating long-range context rather than compressing its content adaptively. Moreover, sequential offline GCRL faces a key bottleneck: under sparse rewards, return-to-go (RTG) becomes non-discriminative across sub-trajectories, providing little guidance signal for stitching goal-reaching behaviors from diverse demonstrations. To address these, we propose \textbf{QHyer}, which replaces RTG with a flow-parameterized, state-conditioned goal-reaching Q-estimator to support stitching across demonstrations, and introduces a gated Hybrid Attention-Mamba backbone that performs content-adaptive history compression while preserving local dynamics. Extensive experiments demonstrate that \textbf{QHyer} achieves state-of-the-art performance on both non-Markovian and Markovian datasets, validating its effectiveness for diverse scenarios.

Xing Lei, Xuetao Zhang, Donglin Wang

Recently, a state-of-the-art family of algorithms, known as Goal-Conditioned Weighted Supervised Learning (GCWSL) methods, has been introduced to tackle challenges in offline goal-conditioned reinforcement learning (RL). GCWSL optimizes a lower bound of the goal-conditioned RL objective and has demonstrated outstanding performance across diverse goal-reaching tasks, providing a simple, effective, and stable solution. However, prior research has identified a critical limitation of GCWSL: the lack of trajectory stitching capabilities. To address this, goal data augmentation strategies have been proposed to enhance these methods. Nevertheless, existing techniques often struggle to sample suitable augmented goals for GCWSL effectively. In this paper, we establish unified principles for goal data augmentation, focusing on goal diversity, action optimality, and goal reachability. Based on these principles, we propose a Model-based Goal Data Augmentation (MGDA) approach, which leverages a learned dynamics model to sample more suitable augmented goals. MGDA uniquely incorporates the local Lipschitz continuity assumption within the learned model to mitigate the impact of compounding errors. Empirical results show that MGDA significantly enhances the performance of GCWSL methods on both state-based and vision-based maze datasets, surpassing previous goal data augmentation techniques in improving stitching capabilities.

Xing Lei, Longjun Liu, Zhiheng Zhou et al.

In this paper, we explore how to design lightweight CNN architecture for embedded computing systems. We propose L-Mobilenet model for ZYNQ based hardware platform. L-Mobilenet can adapt well to the hardware computing and accelerating, and its network structure is inspired by the state-of-the-art work of Inception-ResnetV1 and MobilenetV2, which can effectively reduce parameters and delay while maintaining the accuracy of inference. We deploy our L-Mobilenet model to ZYNQ embedded platform for fully evaluating the performance of our design. By measuring in cifar10 and cifar100 datasets, L-Mobilenet model is able to gain 3x speed up and 3.7x fewer parameters than MobileNetV2 while maintaining a similar accuracy. It also can obtain 2x speed up and 1.5x fewer parameters than ShufflenetV2 while maintaining the same accuracy. Experiments show that our network model can obtain better performance because of the special considerations for hardware accelerating and software-hardware co-design strategies in our L-Mobilenet bottleneck architecture.

Xing Lei, Wenyan Yang, Kaiqiang Ke et al.

Goal-conditioned reinforcement learning (GCRL) with sparse rewards remains a fundamental challenge in reinforcement learning. While hindsight experience replay (HER) has shown promise by relabeling collected trajectories with achieved goals, we argue that trajectory relabeling alone does not fully exploit the available experiences in off-policy GCRL methods, resulting in limited sample efficiency. In this paper, we propose Hindsight Goal-conditioned Regularization (HGR), a technique that generates action regularization priors based on hindsight goals. When combined with hindsight self-imitation regularization (HSR), our approach enables off-policy RL algorithms to maximize experience utilization. Compared to existing GCRL methods that employ HER and self-imitation techniques, our hindsight regularizations achieve substantially more efficient sample reuse and the best performances, which we empirically demonstrate on a suite of navigation and manipulation tasks.

Xing Lei, Zifeng Zhuang, Shentao Yang et al.

Recently, supervised learning (SL) methodology has emerged as an effective approach for offline reinforcement learning (RL) due to their simplicity, stability, and efficiency. However, recent studies show that SL methods lack the trajectory stitching capability, typically associated with temporal difference (TD)-based approaches. A question naturally surfaces: \textit{How can we endow SL methods with stitching capability and close its performance gap with TD learning?} To answer this question, we introduce $Q$-conditioned maximization supervised learning for offline goal-conditioned RL, which enhances SL with the stitching capability through $Q$-conditioned policy and $Q$-conditioned maximization. Concretely, we propose \textbf{G}oal-\textbf{C}onditioned \textbf{\textit{Rein}}forced \textbf{S}upervised \textbf{L}earning (\textbf{GC\textit{Rein}SL}), which consists of (1) estimating the $Q$-function by Normalizing Flows from the offline dataset and (2) finding the maximum $Q$-value within the data support by integrating $Q$-function maximization with Expectile Regression. In inference time, our policy chooses optimal actions based on such a maximum $Q$-value. Experimental results from stitching evaluations on offline RL datasets demonstrate that our method outperforms prior SL approaches with stitching capabilities and goal data augmentation techniques.