Author name cluster

Xu Ma

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

4 papers

1 author row

TAAS Journal 2026 Journal Article

Efficient Resource Allocation for DRL-NOMA Enabled TSIN in Industry 5.0

Yu Cong
Haijun Zhang
Jianquan Wang
Xu Ma
Yan Liang
Yuzheng Ren

To meet sustainable and efficient communication demands of Industry 5.0, this paper proposes an Artificial Intelligence(AI)-driven downlink model for Terrestrial Satellite Integrated Network (TSIN), which integrates both ground and satellite network to serve terrestrial users. Non-Orthogonal Multiple Access (NOMA) improves effectiveness of TSIN, alleviating spectrum resource shortages caused by the explosive growth of User Equipment (UEs). Deep Reinforcement Learning (DRL) is utilized in TSIN to overcome suboptimal convergence. A optimization objective is established with system capacity maximization. The formula is broken down three sub-problems, and a user association scheme is proposed. The process of user association to base stations (BSs) relies on range between BSs and users and channel vectors connecting two. The channel condition ratio is used to determine BS and satellite user sets. NOMA is applied to entire user set. Deep Q-Network (DQN) is used for user grouping, while Deep Deterministic Policy Gradient (DDPG) network is employed for power allocation. The grouping results from DQN network are used as input to DDPG network for cascading training, achieving resource allocation policy optimization. MATLAB simulations show that compared with three typical algorithms, the proposed algorithm improves system capacity and spectral efficiency. This approach provides an efficient communication solution for massive device connectivity of Industry 5.0, effectively promoting sustainable and intelligent development.

Details DOI

NeurIPS Conference 2024 Conference Paper

Slicing Vision Transformer for Flexible Inference

Yitian Zhang
Huseyin Coskun
Xu Ma
Huan Wang
Ke Ma
Xi Chen
Derek H. Hu
Yun Fu

Vision Transformers (ViT) is known for its scalability. In this work, we target to scale down a ViT to fit in an environment with dynamic-changing resource constraints. We observe that smaller ViTs are intrinsically the sub-networks of a larger ViT with different widths. Thus, we propose a general framework, named Scala, to enable a single network to represent multiple smaller ViTs with flexible inference capability, which aligns with the inherent design of ViT to vary from widths. Concretely, Scala activates several subnets during training, introduces Isolated Activation to disentangle the smallest sub-network from other subnets, and leverages Scale Coordination to ensure each sub-network receives simplified, steady, and accurate learning objectives. Comprehensive empirical validations on different tasks demonstrate that with only one-shot training, Scala learns slimmable representation without modifying the original ViT structure and matches the performance of Separate Training. Compared with the prior art, Scala achieves an average improvement of 1. 6% on ImageNet-1K with fewer parameters.

PDF Details DOI

NeurIPS Conference 2022 Conference Paper

Parameter-Efficient Masking Networks

Yue Bai
Huan Wang
Xu Ma
Yitian Zhang
Zhiqiang Tao
Yun Fu

A deeper network structure generally handles more complicated non-linearity and performs more competitively. Nowadays, advanced network designs often contain a large number of repetitive structures (e. g. , Transformer). They empower the network capacity to a new level but also increase the model size inevitably, which is unfriendly to either model restoring or transferring. In this study, we are the first to investigate the representative potential of fixed random weights with limited unique values by learning diverse masks and introduce the Parameter-Efficient Masking Networks (PEMN). It also naturally leads to a new paradigm for model compression to diminish the model size. Concretely, motivated by the repetitive structures in modern neural networks, we utilize one random initialized layer, accompanied with different masks, to convey different feature mappings and represent repetitive network modules. Therefore, the model can be expressed as \textit{one-layer} with a bunch of masks, which significantly reduce the model storage cost. Furthermore, we enhance our strategy by learning masks for a model filled by padding a given random weights vector. In this way, our method can further lower the space complexity, especially for models without many repetitive architectures. We validate the potential of PEMN learning masks on random weights with limited unique values and test its effectiveness for a new compression paradigm based on different network architectures. Code is available at \href{https: //github. com/yueb17/PEMN}{\textcolor{magenta}{https: //github. com/yueb17/PEMN}}.

PDF Details

IJCAI Conference 2020 Conference Paper

Position-Aware Recalibration Module: Learning From Feature Semantics and Feature Position

Xu Ma
Song Fu

We present a new method to improve the representational power of the features in Convolutional Neural Networks (CNNs). By studying traditional image processing methods and recent CNN architectures, we propose to use positional information in CNNs for effective exploration of feature dependencies. Rather than considering feature semantics alone, we incorporate spatial positions as an augmentation for feature semantics in our design. From this vantage, we present a Position-Aware Recalibration Module (PRM in short) which recalibrates features leveraging both feature semantics and position. Furthermore, inspired by multi-head attention, our module is capable of performing multiple recalibrations where results are concatenated as the output. As PRM is efficient and easy to implement, it can be seamlessly integrated into various base networks and applied to many position-aware visual tasks. Compared to original CNNs, our PRM introduces a negligible number of parameters and FLOPs, while yielding better performance. Experimental results on ImageNet and MS COCO benchmarks show that our approach surpasses related methods by a clear margin with less computational overhead. For example, we improve the ResNet50 by absolute 1. 75% (77. 65% vs. 75. 90%) on ImageNet 2012 validation dataset, and 1. 5%~1. 9% mAP on MS COCO validation dataset with almost no computational overhead. Codes are made publicly available.

PDF Details DOI