arXiv subjects

Xing Lei

Publications and source records attributed to Xing Lei.

9 recordsLinked to original sources

arXiv2026

DAGR: State-Conditioned Goal Representations via Difference-Aware Goal Cross-Attention

Goal-conditioned reinforcement learning hinges on how the goal is encoded. Contrastive, metric, temporal-distance, and information-theoretic encoders differ in objective. They still share one trait. None of them sees the current state. Such a state-independent embedding cannot mark which part of the goal still needs action. The policy must then recover that cue by inverting both encoders. We propose DAGR. It refines the static embedding of any late-fusion encoder into a state-conditioned one through multi-scale gated cross-attention. A near-identity gated residual preserves the base representation. Difference-aware Goal Cross-Attention then biases the attention scores using a per-token state-goal discrepancy map. On OGBench, DAGR improves navigation. Our ablations trace the gain to the gated residual, not to the difference bias that names the method. On manipulation and puzzle tasks it matches or falls below the base. DAGR is a structured refinement, not a universal improvement.

cs.LG↗

arXiv2026

NFTR: From Provable Mode-Averaging to Geodesic Subgoal Selection in Offline Goal-Conditioned RL

Hierarchical Implicit Q-Learning (HIQL), an offline goal-conditioned RL method, selects subgoals by value-function advantages alone. This rule has two coupled failure modes. Optimistic bias treats lucky stochastic outcomes as skillful choices, and mode collapse reduces a multi-modal subgoal distribution to a single Gaussian mean that often falls in unreachable regions. We propose NFTR (Normalizing Flows subgoal policies with Triangle-slack Reweighting). A conditional Normalizing Flow replaces the Gaussian policy, and a closed-form mode-averaging result identifies NFs as the minimal generative class for AWR-based subgoal selection. A triangle slack score, built on the architectural triangle inequality without relying on distance accuracy, multiplicatively corrects the AWR weight to downweight subgoals whose detour cost exceeds average reachability. Triangle-slack vanishes on geodesics in deterministic MDPs and remains a conservative upper bound on composability violation under stochastic dynamics. The RWDR objective preserves AWR's population-level monotonic improvement and admits a three-term suboptimality decomposition. Together, these two ingredients yield subgoal selection that provably avoids the Gaussian collapse described above and remains stable under stochastic dynamics. GitHub page: https://github.com/erdemtbao/NFTR

cs.LG↗

arXiv2026

QHyer: Q-conditioned Hybrid Attention-mamba Transformer for Offline Goal-conditioned RL

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.

cs.LG↗

arXiv2025

GCHR : Goal-Conditioned Hindsight Regularization for Sample-Efficient Reinforcement Learning

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.

cs.LG↗

arXiv2025

Closing the Gap between TD Learning and Supervised Learning with $Q$-Conditioned Maximization

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.

cs.LG↗

arXiv2024

MGDA: Model-based Goal Data Augmentation for Offline Goal-conditioned Weighted Supervised Learning

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.

cs.LG↗

arXiv2024

Q-WSL: Optimizing Goal-Conditioned RL with Weighted Supervised Learning via Dynamic Programming

A novel class of advanced algorithms, termed Goal-Conditioned Weighted Supervised Learning (GCWSL), has recently emerged to tackle the challenges posed by sparse rewards in goal-conditioned reinforcement learning (RL). GCWSL consistently delivers strong performance across a diverse set of goal-reaching tasks due to its simplicity, effectiveness, and stability. However, GCWSL methods lack a crucial capability known as trajectory stitching, which is essential for learning optimal policies when faced with unseen skills during testing. This limitation becomes particularly pronounced when the replay buffer is predominantly filled with sub-optimal trajectories. In contrast, traditional TD-based RL methods, such as Q-learning, which utilize Dynamic Programming, do not face this issue but often experience instability due to the inherent difficulties in value function approximation. In this paper, we propose Q-learning Weighted Supervised Learning (Q-WSL), a novel framework designed to overcome the limitations of GCWSL by incorporating the strengths of Dynamic Programming found in Q-learning. Q-WSL leverages Dynamic Programming results to output the optimal action of (state, goal) pairs across different trajectories within the replay buffer. This approach synergizes the strengths of both Q-learning and GCWSL, effectively mitigating their respective weaknesses and enhancing overall performance. Empirical evaluations on challenging goal-reaching tasks demonstrate that Q-WSL surpasses other goal-conditioned approaches in terms of both performance and sample efficiency. Additionally, Q-WSL exhibits notable robustness in environments characterized by binary reward structures and environmental stochasticity.

cs.LG↗

arXiv2024

Exploring Hardware Friendly Bottleneck Architecture in CNN for Embedded Computing Systems

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.

cs.CV↗

arXiv2022

High-performance cavity-enhanced quantum memory with warm atomic cell

High-performance quantum memory for quantized states of light is a prerequisite building block of quantum information technology. Despite great progresses of optical quantum memories based on interactions of light and atoms, physical features of these memories still cannot satisfy requirements for applications in practical quantum information systems, since all of them suffer from trade-off between memory efficiency and excess noise. Here, we report a high-performance cavity-enhanced electromagnetically-induced-transparency memory with warm atomic cell in which a scheme of optimizing the spatial and temporal modes based on the time-reversal approach is applied. The memory efficiency up to 67% is directly measured and a noise level close to quantum noise limit is simultaneously reached. It has been experimentally demonstrated that the average fidelities for a set of input coherent states with different phases and amplitudes within a Gaussian distribution have exceeded the classical benchmark fidelities. Thus the realized quantum memory platform has been capable of preserving quantized optical states, and is ready to be applied in quantum information systems, such as distributed quantum logic gates and quantum-enhanced atomic magnetometry.

quant-ph↗