Ruoning Zhang

Mr. Ruoning Zhang


Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
16 December 2023 in Electronics
Verified
Journal Article
Open access
PDF file
YLYunzhuo Liu
Yunzhuo Liu
RZRuoning Zhang
Ruoning Zhang
SZShijie Zhou
Shijie Zhou
Vehicle simulation algorithms play a crucial role in enhancing traffic efficiency and safety by predicting and evaluating vehicle behavior in various traffic scenarios. Recently, vehicle simulation algorithms based on reinforcement learning have demonstrated excellent performance in practical tasks due to their ability to exhibit superior performance with zero-shot learning. However, these algorithms face challenges in field adaptation problems when deployed in task sets with variable-dimensional observations, primarily due to the inherent limitations of neural network models. In this paper, we propose a neural network structure accommodating variations in specific dimensions to enhance existing reinforcement learning methods. Building upon this, a scene-compatible vehicle simulation algorithm is designed. We conducted experiments on multiple tasks and scenarios using the Highway-Env traffic environment simulator. The results of our experiments demonstrate that the algorithm can successfully operate on all tasks using a neural network model with fixed shape, even with variable-dimensional observations. Our model exhibits no degradation in simulation performance when compared to the baseline algorithm.

27 April 2025 in Mathematics
Verified
Journal Article
Open access
PDF file
ZZZhitong Zhao
Zhitong Zhao
YZYa Zhang
Ya Zhang
SWSiying Wang
Siying Wang
YZYang Zhou
Yang Zhou
RZRuoning Zhang
Ruoning Zhang
WCWenyu Chen
Wenyu Chen
With the development of value decomposition methods, multi-agent reinforcement learning (MARL) has made significant progress in balancing autonomous decision making with collective cooperation. However, the collaborative dynamics among agents are continuously changing. The current value decomposition methods struggle to adeptly handle these dynamic changes, thereby impairing the effectiveness of cooperative policies. In this paper, we introduce the concept of latent interaction, upon which an innovative method for generating weights is developed. The proposed method derives weights from the history information, thereby enhancing the accuracy of value estimations. Building upon this, we further propose a dynamic masking mechanism that recalibrates history information in response to the activity level of agents, improving the precision of latent interaction assessments. Experimental results demonstrate the improved training speed and superior performance of the proposed method in both a multi-agent particle environment and the StarCraft Multi-Agent Challenge.

5 February 2025 in arXiv
Verified
Preprint
Open access
PDF file
YZYang Zhou
Yang Zhou
SWSiying Wang
Siying Wang
WCWenyu Chen
Wenyu Chen
RZRuoning Zhang
Ruoning Zhang
ZZZhitong Zhao
Zhitong Zhao
ZZZixuan Zhang
Zixuan Zhang
Multi-agent reinforcement learning typically employs a centralized training-decentralized execution (CTDE) framework to alleviate the non-stationarity in environment. However, the partial observability during execution may lead to cumulative gap errors gathered by agents, impairing the training of effective collaborative policies. To overcome this challenge, we introduce the Double Distillation Network (DDN), which incorporates two distillation modules aimed at enhancing robust coordination and facilitating the collaboration process under constrained information. The external distillation module uses a global guiding network and a local policy network, employing distillation to reconcile the gap between global training and local execution. In addition, the internal distillation module introduces intrinsic rewards, drawn from state information, to enhance the exploration capabilities of agents. Extensive experiments demonstrate that DDN significantly improves performance across multiple scenarios.

29 October 2026 in Sensors
Verified
Journal Article
Open access
PDF file
CZChao Zhou
Chao Zhou
YGYoucheng Guo
Youcheng Guo
XLXin Li
Xin Li
RZRuoning Zhang
Ruoning Zhang
YHYulin Han
Yulin Han
XCXuexin Chen
Xuexin Chen
In passive electromagnetic detection and positioning systems, the complex electromagnetic environment poses significant theoretical challenges to traditional single-channel blind source separation (SCBSS) algorithms. To address the need to balance separation quality, computational efficiency, resource use and generalisation across different detection conditions, we present the Adaptive-Sampling Mamba–Conv-TasNet, which integrates physically constrained front-end routing with local and long-range sequence modelling. A condition encoder estimates each mixture’s occupied bandwidth and characteristic time scale. It then selects the lowest-rate, anti-aliased profile that preserves the detected band. Four parallel convolutional kernels extract multiscale local features. The separator combines Conv-TasNet-based local modelling with Mamba-based long-range sequence modelling. A shared mask head produces two source-specific masks, and profile-specific decoders subsequently reconstruct two complex-baseband sources. Across three test sets and nine evaluated methods, the proposed model attains the lowest mean complex NMSE on every dataset and the lowest cross-domain log-PSD RMSE while maintaining moderate latency, model size and memory demand. These results indicate that the Adaptive-Sampling Mamba–Conv-TasNet offers a favourable overall trade-off among separation quality, computational efficiency, resource use and cross-domain generalisation relative to the evaluated methods.

5 February 2025 in arXiv
Verified
Preprint
Open access
PDF file
SWSiying Wang
Siying Wang
YZYang Zhou
Yang Zhou
ZZZhitong Zhao
Zhitong Zhao
RZRuoning Zhang
Ruoning Zhang
JSJinliang Shao
Jinliang Shao
YCYuhua Cheng
Yuhua Cheng
Value decomposition (VD) has become one of the most prominent solutions in cooperative multi-agent reinforcement learning. Most existing methods generally explore how to factorize the joint value and minimize the discrepancies between agent observations and characteristics of environmental states. However, direct decomposition may result in limited representation or difficulty in optimization. Orthogonal to designing a new factorization scheme, in this paper, we propose Heterogeneous Policy Fusion (HPF) to integrate the strengths of various VD methods. We construct a composite policy set to select policies for interaction adaptively. Specifically, this adaptive mechanism allows agents' trajectories to benefit from diverse policy transitions while incorporating the advantages of each factorization method. Additionally, HPF introduces a constraint between these heterogeneous policies to rectify the misleading update caused by the unexpected exploratory or suboptimal non-cooperation. Experimental results on cooperative tasks show HPF's superior performance over multiple baselines, proving its effectiveness and ease of implementation.