检索结果-内蒙古大学图书馆

IF2Net: Innately Forgetting-Free Networks for Continual Learning

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Li, Depeng Wang, Tianqi Xu, Bingrong Kawaguchi, Kenji Zeng, Zhigang Suganthan, Ponnuthurai Nagaratnam The School of Artificial Intelligence and Automation The Institute of Artificial Intelligence Huazhong University of Science and Technology The Key Laboratory of Image Processing and Intelligent Control Education Ministry of China Wuhan430074 China The School of Automation Wuhan University of Technology Wuhan430070 China The School of Computing National University of Singapore 117417 Singapore The School of Electrical and Electronic Engineering Nanyang Technological University Singapore639798 Singapore The KINDI Center for Computing Research College of Engineering Qatar University Doha Qatar

Continual learning can incrementally absorb new concepts without interfering with previously learned knowledge. Motivated by the characteristics of neural networks, in which information is stored in weights on connections, we investigated how to design an Innately Forgetting-Free Network (IF2Net) for continual learning context. This study proposed a straightforward yet effective learning paradigm by ingeniously keeping the weights relative to each seen task untouched before and after learning a new task. We first presented the novel representation-level learning on task sequences with random weights. This technique refers to tweaking the drifted representations caused by randomization back to their separate task-optimal working states, but the involved weights are frozen and reused (opposite to well-known layer-wise updates of weights). Then, sequential decision-making without forgetting can be achieved by projecting the output weight updates into the parsimonious orthogonal space, making the adaptations not disturb old knowledge while maintaining model plasticity. IF2Net allows a single network to inherently learn unlimited mapping rules without telling task identities at test time by integrating the respective strengths of randomization and orthogonalization. We validated the effectiveness of our approach in the extensive theoretical analysis and empirical study. Copyright © 2023, The Authors. All rights reserved.

关键词： Decision making

dugMatting: Decomposed-Uncertainty-Guided Matting

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Wu, Jiawei Zhang, Changqing Li, Zuoyong Fu, Huazhu Peng, Xi Zhou, Joey Tianyi College of Mechanical and Electrical Engineering Fujian Agriculture and Forestry University Fuzhou China College of Intelligence and Computing Tianjin University Tianjin China Fujian Provincial Key Laboratory of Information Processing and Intelligent Control Minjiang University Fuzhou China Institute of High Performance Computing Agency for Science Technology and Research Singapore College of Computer Science Sichuan University Chengdu China Singapore

Cutting out an object and estimating its opacity mask, known as image matting, is a key task in image and video editing. Due to the highly ill-posed issue, additional inputs, typically user-defined trimaps or scribbles, are usually needed to reduce the uncertainty. Although effective, it is either time consuming or only suitable for experienced users who know where to place the strokes. In this work, we propose a decomposed-uncertainty-guided matting (dugMatting) algorithm, which explores the explicitly decomposed uncertainties to efficiently and effectively improve the results. Basing on the characteristic of these uncertainties, the epistemic uncertainty is reduced in the process of guiding interaction (which introduces prior knowledge), while the aleatoric uncertainty is reduced in modeling data distribution (which introduces statistics for both data and possible noise). The proposed matting framework relieves the requirement for users to determine the interaction areas by using simple and efficient labeling. Extensively quantitative and qualitative results validate that the proposed method significantly improves the original matting algorithms in terms of both efficiency and efficacy. Copyright © 2023, The Authors. All rights reserved.

关键词： Uncertainty analysis

An advanced actor-critic algorithm for training video game ai 1st

学校读者我要写书评

暂无评论

An advanced actor-critic algorithm for training video game a...

1st International Conference on Neural Computing for Advanced Applications, NCAA 2020

作者： Zha, Zhong Yi Tang, Xue Song Wang, Bo Key Laboratory of Ministry of Education for Image Processing and Intelligent Control Artificial Intelligence and Automation School Huazhong University of Science and Technology Wuhan430074 China College of Information and Science Donghua University Shanghai China

ISBN: (纸本)9789811576690

This paper presents an improved deep reinforcement learning (DRL) algorithm, namely Advanced Actor Critic (AAC), which is based on Actor Critic (AC) algorithm, to the video game Artificial intelligence (AI) training. The advantage distribution estimator, Normal Constraint (NC) function and exploration based on confidence are introduced to improve the AC algorithm, in order to achieve accurate value estimate, select continuous spatial action correctly, and explorate effectively. The aim is to improve the performance of conventional AC algorithm in a complex environment. A case study of video game StarCraft II mini-game AI training is employed. The results verify that the improved algorithm effectively improves the performance in terms of the convergence rate, maximum reward, average reward in every 100 episode, and time to reach a specific reward, etc. The analysis on how these modifications improve the performance is also given through interpretation of the feature layers in the mini-games. © Springer Nature Singapore Pte Ltd. 2020.

关键词： Reinforcement learning

Vision Language Modeling of Content, Distortion and Appearance for Image Quality Assessment

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zhou, Fei Gu, Tianhao Huang, Zhicong Qiu, Guoping College of Electronic and Information Engineering Shenzhen University Shenzhen518060 China Guangdong Provincial Key Laboratory of Intelligent Information Processing Shenzhen518060 China SZU-AFS Joint Innovation Center for Artificial Intelligence Technology Shenzhen518000 China Guangdong-Hong Kong Joint Laboratory for Big Data Imaging and Communication Shenzhen518060 China School of Computer Science The University of Nottingham Ningbo China Zhejiang315100 China School of Computer Science University of Nottingham NottinghamNG8 1BB United Kingdom

The visual quality of an image is confounded by a number of intertwined factors including its semantic content, distortion characteristics and appearance properties such as brightness, contrast, sharpness, and colourfulness. Distilling high level knowledge about all these quality bearing attributes is crucial for developing objective Image Quality Assessment (IQA). While existing solutions have modeled some of these aspects, a comprehensive solution that involves all these important quality related attributes has not yet been developed. In this paper, we present a new blind IQA (BIQA) model termed Self-supervision and Vision-Language supervision Image QUality Evaluator (SLIQUE) that features a joint vision-language and visual contrastive representation learning framework for acquiring high level knowledge about the images semantic contents, distortion characteristics and appearance properties for IQA. For training SLIQUE, we have developed a systematic approach to constructing a first of its kind large image database annotated with all three categories of quality relevant texts. The Text Annotated Distortion, Appearance and Content (TADAC1) database has over 1.6 million images annotated with textual descriptions of their semantic contents, distortion characteristics and appearance properties. The method for constructing TADAC and the database itself will be particularly useful for exploiting vision-language modeling for advanced IQA applications. Extensive experimental results show that SLIQUE has superior performances over state of the art, demonstrating the soundness of its design principle and the effectiveness of its implementation. The link to the TADAC dataset is https://***/documents/tadactextannotated-distortion-appearance-and-content Copyright © 2024, The Authors. All rights reserved.

关键词： Visual languages

Aesthetically Relevant Image Captioning

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Zhong, Zhipeng Zhou, Fei Qiu, Guoping College of Electronics and Information Engineering Shenzhen University China Peng Cheng National Laboratory Shenzhen China Guangdong Key Laboratory of Intelligent Information Processing Shenzhen China Shenzhen Institute for Artificial Intelligence and Robotics for Society China Guangdong-Hong Kong Joint Laboratory for Big Data Imaging and Communication Shenzhen China School of Computer Science The University of Nottingham United Kingdom

Image aesthetic quality assessment (AQA) aims to assign numerical aesthetic ratings to images whilst image aesthetic captioning (IAC) aims to generate textual descriptions of the aesthetic aspects of images. In this paper, we study image AQA and IAC together and present a new IAC method termed Aesthetically Relevant Image Captioning (ARIC). Based on the observation that most textual comments of an image are about objects and their interactions rather than aspects of aesthetics, we first introduce the concept of Aesthetic Relevance Score (ARS) of a sentence and have developed a model to automatically label a sentence with its ARS. We then use the ARS to design the ARIC model which includes an ARS weighted IAC loss function and an ARS based diverse aesthetic caption selector (DACS). We present extensive experimental results to show the soundness of the ARS concept and the effectiveness of the ARIC model by demonstrating that texts with higher ARS’s can predict the aesthetic ratings more accurately and that the new ARIC model can generate more accurate, aesthetically more relevant and more diverse image captions. Furthermore, a large new research database containing 510K images with over 5 million comments and 350K aesthetic scores, and code for implementing ARIC are available at https://***/PengZai/ARIC. © 2022, CC BY-NC-ND.

关键词：

Memristor-based multilayer neural network with edge learning and weight update

学校读者我要写书评

暂无评论

Neurocomputing 2025年 647卷

作者： Ningye Jiang Jupeng Xie Mingxuan Jiang Haoen Huang Zhigang Zeng School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan 430074 China Key Laboratory of Image Information Processing and Intelligent Control Ministry of Education of China Wuhan 430074 China

Memristor crossbar array (MCA) is a promising computing-in-memory architecture for accelerating vector-matrix multiplication (VMM). However, existing memristor-based circuits primarily perform forward computations without local learning capabilities. To address the issue, a memristor-based multilayer neural network (MNN) with edge learning is proposed, enabling error backpropagation and one-step weight update. In particular, the overall circuit design of a three-layer MNN is outlined, comprising three operational stages: forward computation, backpropagation, and weight update. Subsequently, the schematics of four adopted circuit blocks are detailed. By integrating the memory and weight update blocks with MCA, the proposed three-layer memristor-based MNN exhibits scalability and universality, allowing it to be extended to the deep neural network (DNN) architecture. Circuit simulations validate the effectiveness of the proposed scheme, while offline-online co-simulations demonstrate its strength in compensating for variations and non-ideal characteristics of physical devices. Comparative results further highlight the advantages of the proposed scheme in realizing full-analog edge learning circuits and performing error backpropagation processes efficiently.

关键词： Analog circuit design Computing-in-memory Edge learning Memristor crossbar array (MCA) Multilayer neural network

ParamCrop: Parametric cubic cropping for video contrastive learning

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Qing, Zhiwu Huang, Ziyuan Zhang, Shiwei Tang, Mingqian Gao, Changxin Ang, Marcelo H. Jin, Rong Sang, Nong Key Laboratory of Image Processing and Intelligent Control School of Artificial Intelligence and Automation Huazhong University of Science and Technology China ARC National University of Singapore Singapore Alibaba Group China

The central idea of contrastive learning is to discriminate between different instances and force different views from the same instance to share the same representation. To avoid trivial solutions, augmentation plays an important role in generating different views, among which random cropping is shown to be effective for the model to learn a generalized and robust representation. Commonly used random crop operation keeps the distribution of the difference between two views unchanged along the training process. In this work, we show that adaptively controlling the disparity between two augmented views along the training process enhances the quality of the learned representations. Specifically, we present a parametric cubic cropping operation, ParamCrop, for video contrastive learning, which automatically crops a 3D cubic by differentiable 3D affine transformations. ParamCrop is trained simultaneously with the video backbone using an adversarial objective and learns an optimal cropping strategy from the data. The visualizations show that ParamCrop adaptively controls the center distance and the IoU between two augmented views, and the learned change in the disparity along the training process is beneficial to learning a strong representation. Extensive ablation studies demonstrate the effectiveness of the proposed ParamCrop on multiple contrastive learning frameworks and video backbones. Copyright © 2021, The Authors. All rights reserved.

关键词： Crops

Hexagonal Convolutional Neural Network for Spatial Transcriptomics Classification

学校读者我要写书评

暂无评论

Hexagonal Convolutional Neural Network for Spatial Transcrip...

IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

作者： Jing Gao Fa Zhang Kai Hu Xuefeng Cui Key Laboratory of Intelligent Computing and Information Processing of Ministry of Education Xiangtan University Xiangtan China Institute of Computing Technology Chinese Academy of Sciences Beijing China Key Laboratory of Medical Imaging and Artificial Intelligence of Hunan Province Xiangnan University Chenzhou China School of Medical Technology Beijing Institute of Technology Beijing China School of Computer Science and Technology Shandong University Qingdao China

ISBN: (数字)9781665468190

ISBN: (纸本)9781665468206

Recent advances in spatial transcriptomics have enabled the comprehensive measurement of transcriptional profiles while retaining the spatial contextual information. Identifying spatial domains is a critical step in the analysis of spatially resolved transcriptomics. Existing unsupervised methods perform poorly on this task owing to the large amount of noise and dropout events in the transcriptomic profiles. To address this problem, we first extend an unsupervised algorithm to a supervised learning method that can identify useful features and reduce noise hindrance. Second, inspired by the classical convolution in convolutional neural networks (CNNs), we designed a regular hexagonal convolution to compensate for the missing gene expression patterns from adjacent nodes. Compared with the graph convolution in graph neural networks (GNNs), our hexagonal convolution can preserve the relative spatial location information of different nodes in graph-structured data. Third, based on the hexagonal convolution, a novel hexagonal Convolutional Neural Network (hexCNN) is proposed for spatial transcriptomics classification. Finally, we compared the proposed hexCNN with existing methods on the DLPFC dataset. The results show that hexCNN achieves a classification accuracy of 87.2% and an average Rand index (ARI) of 78.2% (1.9% and 3.3% higher than those of GNNs).

关键词： Visualization Convolution Supervised learning Interference Spatial databases Graph neural networks Convolutional neural networks

Fuzzy Sandwich control of Some Nonlinear Systems with a Number of Stochastically Occurring Impulses

学校读者我要写书评

暂无评论

Fuzzy Sandwich Control of Some Nonlinear Systems with a Numb...

International Conference on Information Science and Technology (ICIST)

作者： Babatunde Oluwaseun Onasanya Yuming Feng Xiangguang Dai Wei Zhang School of Computer Science and Engineering Chongqing Three Gorges University Wanzhou Chongqing China Department of Mathematics University of Ibadan Ibadan Nigeria Three Gorges Institute of Artificial Intelligence Chongqing Three Gorges University Wanzhou Chongqing China Chongqing Engineering Research Center of Internet of Things and Intelligent Control Technology Chongqing Three Gorges University Wanzhou Chongqing China Key Laboratory of Intelligent Information Processing and Control Chongqing Three Gorges University Wanzhou Chongqing China

In this paper, rather than have a fixed constant matrix in the sandwich control of the nonlinear system, a fuzzy control matrix was used. In addition, simple inequalities that do not require additional computational time and skills in computer programming were used rather than the usual LMI. The result of this method is that the period T required in the control process is shorter and the control time is also reached much faster or reduced compared to the existing sandwich methods that used constant matrices. In other words, the control time of this method is shorter than the control time achieved when using constant matrices. Besides, this method also can be used to control more nonlinear chaotic systems such as Lorenz's, Rossler's and Chen's systems. Numerical simulations were used to demonstrate the effectiveness of this method.

关键词：