检索结果-内蒙古大学图书馆

Online Attentive Kernel-Based Temporal Difference Learning

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Yang, Guang Chen, Xingguo Yang, Shangdong Wang, Huihui Dong, Shaokang Gao, Yang The the Jiangsu Key Laboratory of Big Data Security & Intelligent Processing Nanjing University of Posts and Telecommunications National Engineering Laboratory for Agri-Product Quality Traceability Beijing Technology and Business University China The State Key Laboratory for Novel Software Technology Nanjing University China The PCA Lab Key Lab of Intelligent Perception and Systems for High-Dimensional Information Ministry of Education Jiangsu Key Lab of Image and Video Understanding for Social Security School of Computer Science and Engineering Nanjing University of Science and Technology China

With rising uncertainty in the real world, online Reinforcement Learning (RL) has been receiving increasing attention due to its fast learning capability and improving data efficiency. However, online RL often suffers from complex Value Function Approximation (VFA) and catastrophic interference, creating difficulty for the deep neural network to be applied to an online RL algorithm in a fully online setting. Therefore, a simpler and more adaptive approach is introduced to evaluate value function with the kernel-based model. Sparse representations are superior at handling interference, indicating that competitive sparse representations should be learnable, non-prior, non-truncated and explicit when compared with current sparse representation methods. Moreover, in learning sparse representations, attention mechanisms are utilized to represent the degree of sparsification, and a smooth attentive function is introduced into the kernel-based VFA. In this paper, we propose an Online Attentive Kernel-Based Temporal Difference (OAKTD) algorithm using two-timescale optimization and provide convergence analysis of our proposed algorithm. Experimental evaluations showed that OAKTD outperformed several Online Kernel-based Temporal Difference (OKTD) learning algorithms in addition to the Temporal Difference (TD) learning algorithm with Tile Coding on public Mountain Car, Acrobot, CartPole and Puddle World tasks. Copyright © 2022, The Authors. All rights reserved.

关键词： Learning algorithms

Taming Consistency Distillation for Accelerated Human image Animation

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Wang, Xiang Zhang, Shiwei Yuan, Hangjie Wei, Yujie Zhang, Yingya Gao, Changxin Wang, Yuehuan Sang, Nong Key Laboratory of Image Processing and Intelligent Control School of Artificial Intelligence and Automation Huazhong University of Science and Technology China Alibaba Group China Fudan University China

Recent advancements in human image animation have been propelled by video diffusion models, yet their reliance on numerous iterative denoising steps results in high inference costs and slow speeds. An intuitive solution involves adopting consistency models, which serve as an effective acceleration paradigm through consistency distillation. However, simply employing this strategy in human image animation often leads to quality decline, including visual blurring, motion degradation, and facial distortion, particularly in dynamic regions. In this paper, we propose the DanceLCM approach complemented by several enhancements to improve visual quality and motion continuity at low-step regime: (1) segmented consistency distillation with an auxiliary light-weight head to incorporate supervision from real video latents, mitigating cumulative errors resulting from single full-trajectory generation;(2) a motion-focused loss to centre on motion regions, and explicit injection of facial fidelity features to improve face authenticity. Extensive qualitative and quantitative experiments demonstrate that DanceLCM achieves results comparable to state-of-the-art video diffusion models with a mere 2-4 inference steps, significantly reducing the inference burden without compromising video quality. The code and models will be made publicly available. Copyright © 2025, The Authors. All rights reserved.

关键词：

A Game Theory-Reinforcement Learning Approach to Cooperation for UAVs

学校读者我要写书评

暂无评论

IEEE Transactions on Vehicular Technology 2025年

作者： Tang, Changbing Pan, Linchao Chen, Jie Liu, Yang Lai, Jingang The College of Physics and Electronic Information Engineering Zhejiang Normal University Jinhua321004 China The School of Mathematical Sciences Zhejiang Normal University Jinhua321004 China The Hangzhou School of Automation Zhejiang Normal University Hangzhou311231 China The College of Electronic Engineering National University of Defense Technology Hefei230037 China The School of Artificial Intelligence and Automation the Key Laboratory of Image Processing and Intelligent Control Education Ministry of China China Huazhong University of Science and Technology Wuhan430074 China

To execute a variety of collaborative tasks, the cooperation for unmanned aerial vehicles (UAVs) with complicated interactions under dynamic environments is a challenging and critical issue. This paper studies the cooperation issue for UAVs under dynamic environments through an approach of game theory-reinforcement learning (GT-RL) approach, which combines the advantages of both game theory and reinforcement learning. First, to cope with the complicated interactions of UAVs, the cluster of UAVs is modeled as a public goods game with asymmetrical environmental feedback. Then, reinforcement learning is adopted to optimize decision-making of UAVs under unknown and dynamic environments through comparing the dynamic behaviors of UAVs, where a novel dynamics system offers a more comprehensive understanding on cooperative behavior among UAVs. Finally, the simulation results show that the GT-RL approach can effectively promote cooperation among UAVs in completing the collaborative tasks. © 1967-2012 IEEE.

关键词： Adversarial machine learning

KinD-LCE: Curve Estimation and Retinex Fusion on Low-Light image

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Lei, Xiaochun Mai, Weiliang Xie, Junlin Liu, He Jiang, Zetao Gong, Zhaoting Lu, Chang Lu, Linjun School of Computer Science and Information Security Guilin University of Electronic Technology Jinji Road Guangxi Guilin541004 China Guangxi Key Laboratory of Image and Graphic Intelligent Processing Guilin University of Electronic Technology Jinji Road Guangxi Guilin541004 China

Low-light images often suffer from noise and color distortion. Object detection, semantic segmentation, instance segmentation, and other tasks are challenging when working with low-light images because of image noise and chromatic aberration. We also found that the conventional Retinex theory loses information in adjusting the image for low-light tasks. In response to the aforementioned problem, this paper proposes an algorithm for low illumination enhancement. The proposed method, KinD-LCE, uses a light curve estimation module to enhance the illumination map in the Retinex decomposed image, improving the overall image brightness. An illumination map and reflection map fusion module were also proposed to restore the image details and reduce detail loss. Additionally, a TV(total variation) loss function was applied to eliminate noise. Our method was trained on the GladNet dataset, known for its diverse collection of low-light images, tested against the Low-Light dataset, and evaluated using the ExDark dataset for downstream tasks, demonstrating competitive performance with a PSNR of 19.7216 and SSIM of 0.8213. Copyright © 2022, The Authors. All rights reserved.

关键词： Semantics

Channel Reflection: Knowledge-Driven Data Augmentation for Eeg-Based Bcis

学校读者我要写书评

暂无评论

SSRN

SSRN 2024年

作者： Wang, Ziwei Li, Siyang Luo, Jingwei Liu, Jiajing Wu, Dongrui Key Laboratory of the Ministry of Education for Image Processing and Intelligent Control School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan430074 China Shenzhen Huazhong University of Science and Technology Research Institute Shenzhen518063 China China Electronic System Technology Co. Ltd. Beijing100089 China School of Civil and Hydraulic Engineering Huazhong University of Science and Technology Wuhan430074 China

A brain-computer interface (BCI) enables direct communication between the human brain and external devices. Electroencephalography (EEG) based BCIs are currently the most popular for able-bodied users. To increase user-friendliness, usually a small amount of user-specific EEG data are used for calibration, which may not be enough to develop a pure data-driven decoding model. To cope with this typical calibration data shortage challenge in EEG-based BCIs, this paper proposes a parameter-free channel reflection (CR) data augmentation approach that incorporates prior knowledge on the channel distributions of different BCI paradigms in data augmentation. Experiments on eight public EEG datasets across four different BCI paradigms (motor imagery, steady-state visual evoked potential, P300, and seizure classifications) using different decoding algorithms demonstrated that: 1) CR is effective, i.e., it can noticeably improve the classification accuracy;2) CR is robust, i.e., it consistently outperforms other data augmentation approaches in the literature;and, 3) CR is flexible, i.e., it can be combined with other data augmentation approaches to further increase the performance. We suggest that data augmentation approaches like CR should be an essential step in EEG-based BCIs. Our code is available online. © 2024, The Authors. All rights reserved.

关键词： Electroencephalography

Named Entity Recognition Based on Anchor Span for Manufacturing Text Knowledge Extraction

学校读者我要写书评

暂无评论

SSRN

SSRN 2024年

作者： Li, Yahui Sun, Qi Zhou, Chunjie Liu, Lu Tian, Yu-Chu School of Artificial Intelligence and Automation Huazhong University of Science and Technology Hubei Wuhan430074 China Key Laboratory of Image Processing and Intelligent Control Ministry of Education Huazhong University of Science and Technology Hubei Wuhan430074 China School of Cyber Science and Engineering Huazhong University of Science and Technology Hubei Wuhan430074 China School of Computer Science Queensland University of Technology BrisbaneQLD4001 Australia

intelligent industrial manufacturing heavily relies on structured knowledge. Named Entity Recognition (NER), an essential technique for extracting structured knowledge from text, has garnered significant research interest. However, current NER approaches face difficulties in handling multiple levels of entity nesting in knowledge extraction, especially within the manufacturing context. To address this issue, we present an Anchor Span-based NER (ASNER) approach for accurately and efficiently extracting manufacturing text knowledge. This method combines token and span classification. Initially, the head and tail features of the entity are extracted, with the corresponding anchor span of potential entities generated based on a boundary match neural network to localize the entities. Subsequently, token spatial contextual features are extracted using biaffine attention and convolutional neural networks, and classification categories are assigned to the anchor span-filtered entity features. Experimental studies demonstrate the effectiveness of the proposed ASNER approach. © 2024, The Authors. All rights reserved.

关键词： Character recognition

Improved black-box attack based on query and perturbation distribution

学校读者我要写书评

暂无评论

Improved black-box attack based on query and perturbation di...

International Workshop on Advanced Computational Intelligence (IWACI)

作者： Weiwei Zhao Zhigang Zeng School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan China Key Laboratory of Image Processing and Intelligent Control of Education Ministry of China Wuhan China

Adversarial examples cause the deep neural network prediction error, which is a great threat to the deep neural network. How to generate more natural adversarial examples and improve the robustness of deep neural networks has received attention. In this paper, we propose an improved blackbox attack (IBBA) algorithm based on query and perturbation distribution. This algorithm only needs the top-l label of the attacked model to generate the adversarial examples. Based on the existing black-box attacks, we optimize the performance of the algorithm from two aspects: query distribution and perturbation distribution. In the aspect of query distribution, we adopt different strategies for nontargeted attack and targeted attack; in the aspect of perturbation distribution, we choose different low-frequency noise according to the difference between the targeted attack and nontargeted attack. The experimental results on imageNet show that the proposed algorithm is better than the existing algorithms in low query number, and the targeted attack is better in each specified query number.

关键词： Perturbation methods Neural networks Prediction algorithms Robustness Low-frequency noise Discrete cosine transforms Computational intelligence

Modality-aware triplet hard mining for zero-shot sketch-based image retrieval

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Huang, Zongheng Sun, Yifan Han, Chuchu Gao, Changxin Sang, Nong Key Laboratory of Ministry of Education for Image Processing and Intelligent Control School of Artificial Intelligence and Automation Huazhong University of Science and Technology China Baidu Research United States

This paper tackles the Zero-Shot Sketch-Based image Retrieval (ZS-SBIR) problem from the viewpoint of cross-modality metric learning. This task has two characteristics: 1) the zero-shot setting requires a metric space with good within-class compactness and the between-class discrepancy for recognizing the novel classes and 2) the sketch query and the photo gallery are in different modalities. The metric learning viewpoint benefits ZS-SBIR from two aspects. First, it facilitates improvement through recent good practices in deep metric learning (DML). By combining two fundamental learning approaches in DML, e.g., classification training and pairwise training, we set up a strong baseline for ZS-SBIR. Without bells and whistles, this baseline achieves competitive retrieval accuracy. Second, it provides an insight that properly suppressing the modality gap is critical. To this end, we design a novel method named Modality-Aware Triplet Hard Mining (MATHM). MATHM enhances the baseline with three types of pairwise learning, e.g., a cross-modality sample pair, a within-modality sample pair, and their combination. We also design an adaptive weighting method to balance these three components during training dynamically. Experimental results confirm that MATHM brings another round of significant improvement based on the strong baseline and sets up new state-of-the-art performance. For example, on the TU-Berlin dataset, we achieve 47.88+2.94% mAP@all and 58.28+2.34% Prec@100. Code will be publicly available. Copyright © 2021, The Authors. All rights reserved.

关键词： Deep learning

SiamCMN: Jointing cycle memory and correlation network for Siamese object tracking

学校读者我要写书评

暂无评论

Optik 2023年第1期286卷

作者： Zhang, Baohua Zhang, Nianchao Li, Yongxiang Lu, Xiaoqi Gu, Yu Li, Jianjun School of Information Engineering Inner Mongolia University of Science and Technology Baotou Inner Mongolia014010 China Inner Mongolia Key Laboratory of Pattern Recognition and Intelligent Image Processing Baotou Inner Mongolia014010 China School of Information Engineering Mongolia Industrial University Hohhot Inner Mongolia010051 China College of Energy and Transportation Engineering Inner Mongolia Agricultural University Hohhot Inner Mongolia010018 China

In heterogeneous scenes with latent non-deterministic states, template features determine the performance of the Siamese network-based trackers, however, background noise is easily introduced in the search area matching process, which causes the trackers lack sufficient discrimination. To solve this problem, this paper proposes an object tracking algorithm based on the cyclic memory module. Firstly, the extracted primary features are mapped to the cyclic spatial memory network, the spatial memory unit is used to store the interference components. On this basis, the cyclic structure is used to filter out the interference components and refine the template features. Then a channel correlation network is introduced into the proposed method, which can mine semantic complementary information and capture global contextual relevance by modeling the positive and the negative correlation between the channels. The experimental results on tracking benchmarks such as OTB50, OTB100, UAV123, VOT2018, GOT10k, and LaSOT have shown that the proposed method outperforms recent representative algorithms in tracking accuracy in challenging scenarios. © 2023 Elsevier GmbH

关键词： Semantics