检索结果-内蒙古大学图书馆

Self-Supervised Interactive Embedding for One-Shot Organ Segmentation

IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING 2023年第10期70卷 2799-2808页

作者： Yang, Yang Wang, Bo Zhang, Dingwen Yuan, Yixuan Yan, Qingsen Zhao, Shijie You, Zheng Han, Junwei Northwestern Polytech Univ Sch Automat Xian Peoples R China Tsinghua Univ Dept Precis Instrument State Key Lab Precis Measurement Technol & Instrum Beijing Peoples R China Technology Ltd Beijing Jingzhen Med Beijing Peoples R China Fourth Mil Med Univ Xijing Hosp Dept Clin Immunol Xian 710032 Peoples R China Hefei Comprehens Natl Sci Ctr Inst Artificial Intelligence Hefei 230088 Peoples R China Northwestern Polytech Univ Sch Automat Brain & Artificial Intelligence Lab Xian 710072 Peoples R China Chinese Univ Hong Kong Dept Elect Engn Hong Kong Peoples R China Northwestern Polytech Univ Sch Comp Sci Xian Peoples R China Tsinghua Univ Dept Precis Instrument State Key Lab Precis Measurement Technol & Instrum Beijing Peoples R China

One-shot organ segmentation (OS2) aims at segmenting the desired organ regions from the input medical imaging data with only one pre-annotated example as the reference. By using the minimal annotation data to facilitate organ segmentation, OS2 receives great attention in the medical image analysis community due to its weak requirement on human annotation. In OS2, one core issue is to explore the mutual information between the support (reference slice) and the query (test slice). Existing methods rely heavily on the similarity between slices, and additional slice allocation mechanisms need to be designed to reduce the impact of the similarity between slices on the segmentation performance. To address this issue, we build a novel support-query interactive embedding (SQIE) module, which is equipped with the channel-wise co-attention, spatial-wise co-attention, and spatial bias transformation blocks to identify "what to look", "where to look", and "how to look" in the input test slice. By combining the three mechanisms, we can mine the interactive information of the intersection area and the disputed area between slices, and establish the feature connection between the target in slices with low similarity. We also propose a self-supervised contrastive learning framework, which transforms knowledge from the physical position to the embedding space to facilitate the self-supervised interactive embedding of the query and support slices. Comprehensive experiments on two large benchmarks demonstrate the superior capacity of the proposed approach when compared with the current alternatives and baseline models.

关键词： Image segmentation Biomedical imaging Three-dimensional displays Task analysis Annotations Feature extraction Biological systems Self-supervised learning contrastive learning interactive embedding co-attention mechanism one-shot learning medical image segmentation

来源：评论

学校读者我要写书评

暂无评论

Distributed Prescribed-Time Convex Optimization: Cascade Design and Time-Varying Gain Approach

arXiv

引用

arXiv 2024年

作者： Zuo, Gewei Zhu, Lijun Wang, Yujuan Chen, Zhiyong School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan430072 China Key Laboratory of Imaging Processing and Intelligence Control Huazhong University of Science and Technology Wuhan430074 China The State Key Laboratory of Power Transmission Equipment & System Security and New Technology School of Automation Chongqing University Chongqing400044 China School of Engineering The University of Newcastle CallaghanNSW2308 Australia

In this paper, we address the distributed prescribed-time convex optimization (DPTCO) problem for a class of high-order nonlinear multi-agent systems (MASs) under undirected connected graphs. A cascade design framework is proposed, dividing the DPTCO implementation into two parts: distributed optimal trajectory generator design and local reference trajectory tracking controller design. The DPTCO problem is then transformed into the prescribed-time stabilization problem of a cascaded system. Changing Lyapunov function and time-varying state transformation methods together with the sufficient conditions are proposed to prove the prescribed-time stabilization of the cascaded system as well as the uniform boundedness of internal signals in the closed-loop MASs. The proposed framework is then utilized to solve robust DPTCO problem for a class of chain-integrator MASs with external disturbances by constructing a novel sliding-mode variables and exploiting the property of time-varying gains. The proposed framework is further utilized to solve the adaptive DPTCO problem for a class of strict-feedback MASs with parameter uncertainty, in which backstepping method with prescribed-time dynamic filter is adopted. The descending power state transformation is introduced to compensate the growth of increasing rate induced by the derivative of time-varying gains in recursive steps and the high-order derivative of local reference trajectory is not required. Finally, theoretical results are verified by two numerical examples. Copyright © 2024, The Authors. All rights reserved.

关键词： Stabilization

来源：评论

学校读者我要写书评

暂无评论

Simplified Neural Network Based Optical Performance Monitoring Using Amplitude Histogram Metrics and Tap Coefficients 29

Simplified Neural Network Based Optical Performance Monitori...

引用

29th Opto-Electronics and Communications Conference, OECC 2024

作者： Wan, Zhiquan Yu, Zhenming Xu, Kun Yin, Kun Yu, Hui Hu, Chuliang Zhejiang Lab Research Center of High Efficiency Computing System Hangzhou China Beijing University of Posts and Telecommunications State Key Laboratory of Information Photonics and Optical Communications Beijing China Chinese Academy of Sciences Institute of Innovative Computing Technology Hangzhou China

We demonstrate a simplified neural network to estimate OSNR in nonlinear regime using tap coefficients and metrics extracted from amplitude histogram. The MSE of OSNR estimation improves 1 dB with the help of tap coef... 详细信息

ISBN: (纸本)9798350308396

关键词： adaptive equalization algorithm neural network optical performance monitoring OSNR

来源：评论

学校读者我要写书评

暂无评论

U-REPA: Aligning Diffusion U-Nets to ViTs

arXiv

引用

arXiv 2025年

作者： Tian, Yuchuan Chen, Hanting Zheng, Mengyu Liang, Yuchen Xu, Chao Wang, Yunhe State Key Lab of General AI School of Intelligence Science and Technology Peking University China Huawei Noah’s Ark Lab Canada The University of Sydney Australia School of Mathematical Sciences Peking University China

Representation Alignment (REPA) that aligns Diffusion Transformer (DiT) hidden-states with ViT visual encoders has proven highly effective in DiT training, demonstrating superior convergence properties, but it has not been validated on the canonical diffusion U-Net architecture that shows faster convergence compared to DiTs. However, adapting REPA to U-Net architectures presents unique challenges: (1) different block functionalities necessitate revised alignment strategies;(2) spatial-dimension inconsistencies emerge from U-Net’s spatial downsampling operations;(3) space gaps between U-Net and ViT hinder the effectiveness of tokenwise alignment. To encounter these challenges, we propose U-REPA, a representation alignment paradigm that bridges U-Net hidden states and ViT features as follows: Firstly, we propose via observation that due to skip connection, the middle stage of U-Net is the best alignment option. Secondly, we propose upsampling of U-Net features after passing them through MLPs. Thirdly, we observe difficulty when performing tokenwise similarity alignment, and further introduces a manifold loss that regularizes the relative similarity between samples. Experiments indicate that the resulting U-REPA could achieve excellent generation quality and greatly accelerates the convergence speed. With CFG guidance interval, U-REPA could reach FID Copyright © 2025, The Authors. All rights reserved.

关键词： Alignment

来源：评论

学校读者我要写书评

暂无评论

Self-Supervised Teaching and Learning of Representations on Graphs 23

Self-Supervised Teaching and Learning of Representations on ...

引用

32nd ACM World Wide Web Conference, WWW 2023

作者： Wan, Liangtian Fu, Zhenqiang Sun, Lu Wang, Xianpeng Xu, Gang Yan, Xiaoran Xia, Feng Key Laboratory for Ubiquitous Network Service Software of Liaoning Province School of Software Dalian University of Technology Dalian China Department of Communication Engineering Institute of Information Science Technology Dalian Maritime University Dalian China State Key Laboratory of Marine Resource Utilization in South China Sea School of Information and Communication Engineering Hainan University Haikou China State Key Laboratory of Millimeter Waves School of Information Science and Engineering Southeast University Nanjing China Research Center of Big Data Intelligence Research Institute of Artificial Intelligence Zhejiang Lab Hangzhou China School of Computing Technologies Rmit University Melbourne Australia

ISBN: (纸本)9781450394161

Recent years have witnessed significant advances in graph contrastive learning (GCL), while most GCL models use graph neural networks as encoders based on supervised learning. In this work, we propose a novel graph learning model called GraphTL, which explores self-supervised teaching and learning of representations on graphs. One critical objective of GCL is to retain original graph information. For this purpose, we design an encoder based on the idea of unsupervised dimensionality reduction of locally linear embedding (LLE). Specifically, we map one iteration of the LLE to one layer of the network. To guide the encoder to better retain the original graph information, we propose an unbalanced contrastive model consisting of two views, which are the learning view and the teaching view, respectively. Furthermore, we consider the nodes that are identical in muti-views as positive node pairs, and design the node similarity scorer so that the model can select positive samples of a target node. Extensive experiments have been conducted over multiple datasets to evaluate the performance of GraphTL in comparison with baseline models. Results demonstrate that GraphTL can reduce distances between similar nodes while preserving network topological and feature information, yielding better performance in node classification. © 2023 ACM.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

Beyond Textual Constraints: Learning Novel Diffusion Conditions with Fewer Examples

Beyond Textual Constraints: Learning Novel Diffusion Conditi...

引用

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Yuyang Yu Bangzhen Liu Chenxi Zheng Xuemiao Xu Shengfeng He Huaidong Zhang South China University of Technology State Key Laboratory of Subtropical Building Science Guangdong Provincial Key Lab of Computational Intelligence and Cyberspace Information Ministry of Education Key Laboratory of Big Data and Intelligent Robot Singapore Management University

ISBN: (数字)9798350353006

ISBN: (纸本)9798350353013

In this paper, we delve into a novel aspect of learning novel diffusion conditions with datasets an order of magnitude smaller. The rationale behind our approach is the elimination of textual constraints during the few-shot learning process. To that end, we implement two optimization strategies. The first, prompt-free conditional learning, utilizes a prompt-free encoder derived from a pre-trained Stable Diffusion model. This strategy is designed to adapt new conditions to the diffusion process by minimizing the textual-visual cor-relation, thereby ensuring a more precise alignment between the generated content and the specified conditions. The second strategy entails condition-specific negative rectification, which addresses the inconsistencies typically brought about by Classifier-free guidance in few-shot training con-texts. Our extensive experiments across a variety of condition modalities demonstrate the effectiveness and efficiency of our framework, yielding results comparable to those obtained with datasets a thousand times larger. Our codes are available at https://***/Yuyan9Yu/BeyondTextConstraint.

关键词： Training Computer vision Adaptation models Codes Text to image Diffusion processes Diffusion models

来源：评论

学校读者我要写书评

暂无评论

Beyond Simple Sum of Delayed Rewards: Non-Markovian Reward Modeling for Reinforcement Learning

arXiv

引用

arXiv 2024年

作者： Tang, Yuting Cai, Xin-Qiang Pang, Jing-Cheng Wu, Qiyu Ding, Yao-Xiang Sugiyama, Masashi The University of Tokyo Japan RIKEN Center for Advanced Intelligence Project Japan National Key Laboratory for Novel Software Technology Nanjing University China State Key Lab for CAD & CG Zhejiang University China

Reinforcement Learning (RL) empowers agents to acquire various skills by learning from reward signals. Unfortunately, designing high-quality instance-level rewards often demands significant effort. An emerging alternative, RL with delayed reward, focuses on learning from rewards presented periodically, which can be obtained from human evaluators assessing the agent’s performance over sequences of behaviors. However, traditional methods in this domain assume the existence of underlying Markovian rewards and that the observed delayed reward is simply the sum of instance-level rewards, both of which often do not align well with real-world scenarios. In this paper, we introduce the problem of RL from Composite Delayed Reward (RLCoDe), which generalizes traditional RL from delayed rewards by eliminating the strong assumption. We suggest that the delayed reward may arise from a more complex structure reflecting the overall contribution of the sequence. To address this problem, we present a framework for modeling composite delayed rewards, using a weighted sum of non-Markovian components to capture the different contributions of individual steps. Building on this framework, we propose Composite Delayed Reward Transformer (CoDeTr), which incorporates a specialized in-sequence attention mechanism to effectively model these contributions. We conduct experiments on challenging locomotion tasks where the agent receives delayed rewards computed from composite functions of observable step rewards. The experimental results indicate that CoDeTr consistently outperforms baseline methods across evaluated metrics. Additionally, we demonstrate that it effectively identifies the most significant time steps within the sequence and accurately predicts rewards that closely reflect the environment feedback. Code is available at an anonymous link: https://***/r/CoDe-67E8/. © 2024, CC BY-NC-ND.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Gait recognition based on sEMG and Deep Residual Shrinkage Network

Gait recognition based on sEMG and Deep Residual Shrinkage N...

引用

IEEE International Conference on Real-time Computing and Robotics (RCAR)

作者： Mingxiang Luo Lijun Yang Zeyu Sun Meng Yin Yue Ma Xinyu Wu Wujing Cao Shenzhen Institute of Advanced Technology Guangdong Provincial Key Lab of Robotics and Intelligent System Chinese Academy of Sciences and Southern University of Science and Technology Shenzhen Institute of Advanced Technology Guangdong Provincial Key Lab of Robotics and Intelligent System Chinese Academy of Sciences Shenzhen China SIAT Branch Shenzhen Institute of Artificial Intelligence and Robotics for Society University of Birmingham

Transverse resistance exoskeleton is expected to provide better rehabilitation training methods for patients with lower limb injury. Lateral walking gait recognition plays an important role in lateral resistance exoskeleton control. In the phase of lateral walking out of synchrony, the EMG signals produced by the thighs will be different. The EMG signals collected by five subjects while walking laterally were used in the experiment. The four muscles are: the gluteus medius of the left leg, the tensor fascia lata, the adductor major, and the tensor fascia lata of the right leg. Based on the one-dimensional EMG signals of the four muscles, this paper uses the classification model commonly used in the MATlab toolbox for training and recognition. However, the experimental results show that TREE and KNN have poor recognition accuracy for the four gait stages. To this end, this paper applies the convolutional residual shrinkage network (DRSN) to this occasion. The collected one-dimensional EMG signals are plotted as pictures and fed into the model for training and classification. Experimental results show that the classification accuracy is greatly improved compared with the previous model.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Safe reinforcement learning using finite-horizon gradient-based estimation 24

Safe reinforcement learning using finite-horizon gradient-ba...

引用

Proceedings of the 41st International Conference on Machine Learning

作者： Juntao Dai Yaodong Yang Qian Zheng Gang Pan College of Computer Science and Technology and The State Key Lab of Brain-Machine Intelligence Zhejiang University Hangzhou China Center for AI Safety and Governance Peking University Beijing China

A key aspect of Safe Reinforcement Learning (Safe RL) involves estimating the constraint condition for the next policy, which is crucial for guiding the optimization of safe policy updates. However, the existing Advantage-based Estimation (ABE) method relies on the infinite-horizon discounted advantage function. This dependence leads to catastrophic errors in finite-horizon scenarios with non-discounted constraints, resulting in safety-violation updates. In response, we propose the first estimation method for finite-horizon non-discounted constraints in deep Safe RL, termed Gradient-based Estimation (GBE), which relies on the analytic gradient derived along trajectories. Our theoretical and empirical analyses demonstrate that GBE can effectively estimate constraint changes over a finite horizon. Constructing a surrogate optimization problem with GBE, we developed a novel Safe RL algorithm called Constrained Gradient-based Policy Optimization (CGPO). CGPO identifies feasible optimal policies by iteratively resolving sub-problems within trust regions. Our empirical results reveal that CGPO, unlike baseline algorithms, successfully estimates the constraint functions of subsequent policies, thereby ensuring the efficiency and feasibility of each update.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Richelieu: Self-Evolving LLM-Based Agents for AI Diplomacy

arXiv

引用

arXiv 2024年

作者： Guan, Zhenyu Kong, Xiangyu Zhong, Fangwei Wang, Yizhou Institute for Artificial Intelligence Peking University Beijing China Computer School Beijing Information Science & Technology University Beijing China School of Artificial Intelligence Beijing Normal University Beijing China Center on Frontiers of Computing Studies School of Computer Science Nat’l Eng. Research Center of Visual Technology State Key Lab of General Artificial Intelligence Peking University Beijing China State Key Laboratory of General Artificial Intelligence BIGAI Beijing China

Diplomacy is one of the most sophisticated activities in human society, involving complex interactions among multiple parties that require skills in social reasoning, negotiation, and long-term strategic planning. Previous AI agents have demonstrated their ability to handle multi-step games and large action spaces in multi-agent tasks. However, diplomacy involves a staggering magnitude of decision spaces, especially considering the negotiation stage required. While recent agents based on large language models (LLMs) have shown potential in various applications, they still struggle with extended planning periods in complex multi-agent settings. Leveraging recent technologies for LLM-based agents, we aim to explore AI’s potential to create a human-like agent capable of executing comprehensive multi-agent missions by integrating three fundamental capabilities: 1) strategic planning with memory and reflection;2) goal-oriented negotiation with social reasoning;and 3) augmenting memory through self-play games for self-evolution without human in the loop. Copyright © 2024, The Authors. All rights reserved.

关键词： Strategic planning

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：