检索结果-内蒙古大学图书馆

Multi-agent Dynamic Algorithm Configuration

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Xue, Ke Xu, Jiacheng Yuan, Lei Li, Miqing Qian, Chao Zhang, Zongzhang Yu, Yang State Key Laboratory for Novel Software Technology Nanjing University China Polixir Technologies China CERCIA School of Computer Science University of Birmingham United Kingdom

Automated algorithm configuration relieves users from tedious, trial-and-error tuning tasks. A popular algorithm configuration tuning paradigm is dynamic algorithm configuration (DAC), in which an agent learns dynamic configuration policies across instances by reinforcement learning (RL). However, in many complex algorithms, there may exist different types of configuration hyperparameters, and such heterogeneity may bring difficulties for classic DAC which uses a single-agent RL policy. In this paper, we aim to address this issue and propose multi-agent DAC (MA-DAC), with one agent working for one type of configuration hyperparameter. MA-DAC formulates the dynamic configuration of a complex algorithm with multiple types of hyperparameters as a contextual multi-agent Markov decision process and solves it by a cooperative multi-agent RL (MARL) algorithm. To instantiate, we apply MA-DAC to a well-known optimization algorithm for multi-objective optimization problems. Experimental results show the effectiveness of MA-DAC in not only achieving superior performance compared with other configuration tuning approaches based on heuristic rules, multi-armed bandits, and single-agent RL, but also being capable of generalizing to different problem classes. Furthermore, we release the environments in this paper as a benchmark for testing MARL algorithms, with the hope of facilitating the application of MARL. Copyright © 2022, The Authors. All rights reserved.

关键词： Reinforcement learning

MII:A Novel Text Classification Model Combining Deep Active Learning with BERT

学校读者我要写书评

暂无评论

computers, Materials & Continua 2020年第6期63卷 1499-1514页

作者： Anman Zhang Bohan Li Wenhuan Wang Shuo Wan Weitong Chen College of Computer Science and Technology Nanjing University of Aeronautics and AstronauticsNanjing211106China Key Laboratory of Safety-Critical Software Ministry of Industry and Information TechnologyNanjing211106China Collaborative Innovation Center of Novel Software Technology and Industrialization Nanjing210046China School of Information Technology and Electrical Engineering University of QueenslandQueenslandAustralia

Active learning has been widely utilized to reduce the labeling cost of supervised *** selecting specific instances to train the model,the performance of the model was improved within limited ***,rare work paid attention to the effectiveness of active learning on *** this paper,we proposed a deep active learning model with bidirectional encoder representations from transformers(BERT)for text *** takes advantage of the self-attention mechanism to integrate contextual information,which is beneficial to accelerate the convergence of *** for the process of active learning,we design an instance selection strategy based on posterior probabilities Margin,Intra-correlation and Inter-correlation(MII).Selected instances are characterized by small margin,low intra-cohesion and high *** conduct extensive experiments and analytics with our *** effect of learner is compared while the effect of sampling strategy and text classification is assessed from three real *** results show that our method outperforms the baselines in terms of accuracy.

关键词： Active learning instance selection deep neural network text classification

SARA: Controllable Makeup Transfer with Spatial Alignment and Region-Adaptive Normalization

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Zhong, Xiaojing Huang, Xinyi Wu, Zhonghua Lin, Guosheng Wu, Qingyao School of Software Engineering South China University of Technology China School of Computer Science and Engineering Nanyang Technological University Singapore Key Laboratory of Big Data and Intelligent Robot Ministry of Education China Peng Cheng Laboratory China SenseTime Research China

Makeup transfer is a process of transferring the makeup style from a reference image to the source image, while preserving the source image’s identity. In addition to achieving fine-grained control over the makeup transfer, incorporating semantic alignment into the transformation is crucial, as the poses of the reference and source images are often inconsistent. We propose a novel Semantic-guided Alignment and Region-Adaptive normalization framework (SARA) to effectively transfer makeup styles under misaligned poses, offering flexible control to meet the demands of real-world applications, such as partial transfer, intensity adjustment, and makeup removal. Specifically, SARA comprises three modules: Firstly, we propose a semantic-guided alignment module to explicitly construct dense correspondence between the reference image and the target semantic map, employing unbalanced optimal transport matching to handle semantic region mismatches. Secondly, a region-adaptive normalization module is responsible for dynamically combining the warped style features with shape-independent style codes obtained by region-wise average pooling, mitigating feature loss during the alignment. Lastly, a makeup fusion module progressively fuses the identity features with the makeup styles to render the final output image. Furthermore, we combine optimal transport with histogram matching to generate the pseudo ground truth, which is used to facilitate the transfer in terms of both spatial alignment and color distribution. Experimental results show that our proposed SARA outperforms existing methods on two public datasets. © 2023, CC BY.

关键词： Semantics

A Robust Methodology for Fruit Quality Prediction and Estimation Using Intelligent Learning Based Image Processing Logic

学校读者我要写书评

暂无评论

A Robust Methodology for Fruit Quality Prediction and Estima...

Innovative Computing, Intelligent Communication and Smart Electrical Systems (ICSES), International Conference on

作者： Malathi K Vunnava Dinesh Babu M. Ramkumar M. Ayyadurai Vivek K Hamzeh J Aljawawdeh Department of CSE Saveetha School of Engineering Saveetha Institute of Science and Technology (SIMATS) Saveetha University Chennai Department of CSE RV Institute of Technology Guntur Department of Computer Science and Engineering Rajalakshmi Engineering College Chennai Chennai Department of Mechanical Engineering Yeshwantrao Chavan College of Engineering Nagpur Department of software engineering Faculty of information technology Zarqa University Zarqa Jordan University of Business and Technology Jeddah Saudi Arabia

ISBN: (数字)9798331543624

ISBN: (纸本)9798331543631

Fruit quality prediction plays a vital role in agricultural and food industries, ensuring that only high-quality produce reaches the consumer. Traditional methods of assessing fruit quality rely on manual inspection, which is both time-consuming and prone to human error. This paper presents a robust methodology for fruit quality prediction and estimation, utilizing a hybrid deep learning model that integrates DenseN et for feature extraction with LightGBM for efficient classification. The proposed model is designed to automate the process of fruit quality assessment by analyzing key attributes such as color, texture, and shape from high-resolution fruit images. The Fruit360 dataset was used for training and testing, with images classified into categories such as fresh, ripe, overripe, and damaged. A comprehensive comparison with nine existing models, including SVM, Random Forest, and XGBoost, was conducted using performance metrics such as accuracy, precision, recall, and Fl-score. The proposed model achieved an accuracy of 96.78%, significantly outperforming all other models. Furthermore, the model's precision and recall reached 95.33% and 95.67%, respectively, demonstrating its reliability in distinguishing high-quality fruits from low-quality ones.

关键词： Training Support vector machines Accuracy Shape Computational modeling Estimation Predictive models Feature extraction Reliability Testing

Flexi-Compression: A Flexible Model Compression Method for Autonomous Driving 11

学校读者我要写书评

暂无评论

Flexi-Compression: A Flexible Model Compression Method for A...

11th ACM Symposium on Design and Analysis of Intelligent Vehicular Networks and Applications, DIVANet 2021

作者： Liu, Hantao He, Ying Richard Yu, F. James, Jeremy School of Information Technology Carleton University OttawaON Canada College of Computer Science and Software Engineering Shenzhen University Shenzhen Guangzhou China Carleton University OttawaON Canada Four DRobotics Corp OttawaON Canada

ISBN: (纸本)9781450390811

Benefiting from the rapid development of convolutional neural networks, computer vision-based autonomous driving technologies are gradually being deployed in vehicles. However, these neural networks typically have a large number of parameters and extremely high computational cost, making them difficult to deploy in autonomous vehicles with limited storage and computational power. In this paper, we propose an innovative model compression approach to compress convolutional neural networks in autonomous driving algorithms, which we call Flexi-Compression. Flexi-Compression first modifies the model structure by replacing the traditional convolutional layers with our proposed Flexi-CP module, thus reducing the computation of the convolutional layers. Then, we leverage knowledge distillation to enable the compressed model to quickly acquire the knowledge of the original model. In addition, we use a Flexi-Batch Normalization layer to prune the model and finally further reduce the model size by model quantization. We compress the VGG-16 network using our proposed model compression algorithm, which is a commonly used backbone network in autonomous driving algorithms. On the CIFAR-10 dataset, our compression method can reduce the parameters of the VGG-16 network by 86% and the computation by 87% with 4% loss of accuracy. To verify the effectiveness of our compression algorithm in real-world applications, we also compress an autonomous driving algorithm and achieve excellent performance. © 2021 ACM.

关键词： Convolution

Robust Multi-View Prototype Learning

学校读者我要写书评

暂无评论

Robust Multi-View Prototype Learning

2020 International Conference on Internet of Things and Intelligent Applications, ITIA 2020

作者： Tian, Qing Xia, Shiyu Cao, Meng Cheng, Keyang Nanjing University of Information Science and Technology School of Computer and Software Nanjing China Southeast University School of Computer Science and Engineering Nanjing China Nanjing University of Aeronautics and Astronautics College of Computer Science and Technology Nanjing China Jiangsu University School of Computer Science and Communication Engineering Zhenjiang China

ISBN: (纸本)9781728193014

In vision and machine learning, information fusion from multiple sensors can be regarded as multi-view learning paradigm to make use of the pairwise complementary information. Due to disturbed variances by illumination, equipment and environment, the collected data is frequently smeared by noises. Although there have been outlier-against works proposed, most of them suffer from redundant parameters. Even worse, few of them have modeled the cross-view complementary information. Actually, in multi-view learning, not only the clean data points, but also those outliers share the same prototypes in common space, by contrast, the margin value of the latter is non-zero. Motivated by this, in this article, we construct a robust multi-view prototype (RMVP) learning model with auxiliary margin matrices. Specifically, in RMVP, the projection matrices corresponding to each view span the common space to characterize the prototypes, simultaneously. In addition, an auxiliary sparse margin matrix for corresponding view identifies the distance gaps of outliers, whose non-zero column vector represents the margin value. To improve the robustness, the lasso regularization is penalized on the auxiliary matrices. And, an alternating optimization algorithm is presented to solve the proposed model. Finally, extensive experiments are constructed to testify the effectiveness of the proposed method. © 2020 IEEE.

关键词： Matrix algebra

FedMDC: Enabling Communication-Efficient Federated Learning over Packet Lossy Networks via Multiple Description Coding

学校读者我要写书评

暂无评论

FedMDC: Enabling Communication-Efficient Federated Learning ...

IEEE International Conference on Multimedia and Expo (ICME)

作者： Yixuan Guan Xuefeng Liu Tao Ren Jianwei Niu State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and Engineering Beihang University Beijing China Zhongguancun Laboratory Beijing China State Key Laboratory of Intelligent Game Institute of Software Chinese Academy of Sciences Beijing China Zhengzhou University Research Institute of Industrial Technology School of Information Engineering Zhengzhou University Zhengzhou China

ISBN: (数字)9798350390155

ISBN: (纸本)9798350390162

Federated learning (FL) generally suffers significant communication overhead from high-traffic gradient synchronization. The majority of existing studies on this problem aim at compressing gradients under the premise of reliable transmission. While transmission reliability can be ensured via TCP by default, the notably increased latency and retransmitted packets are prohibitive for most clients in FL. To tackle this issue, we propose FedMDC, a retransmission-free compression framework for FL over packet lossy networks. Given clients’ limited resources, FedMDC adopts multiple description coding to encode gradients into redundant descriptions for erasure resilience simply through multiplying an overcomplete matrix; and then quantizes these descriptions for compression. To further reduce quantization distortion and computational overhead, a reduced decoding algorithm is developed by decoding the aggregation of all clients’ encodings in conjunction with a customized dither quantization design. Besides, FedMDC explicitly supports adaptive bitrates subject to clients’ heterogeneous communication budgets, which maximize resource utilization to facilitate distortion reduction and accelerate model convergence. Theoretical analysis and experimental results both demonstrate the effectiveness of our scheme.

关键词： Quantization (signal) Federated learning Distortion Propagation losses Encoding Decoding Reliability Synchronization Resource management Resilience

Semantic-Constraint Matching Transformer for Weakly Supervised Object Localization

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Cao, Yiwen Su, Yukun Wang, Wenjun Liu, Yanxia Wu, Qingyao The School of Software Engineering South China University of Technology Guangzhou510640 China The Key Laboratory of Big Data and Intelligent Robot Ministry of Education Guangzhou China The School of Computer Science and Engineering Nanyang Technological University Singapore The Pazhou Lab Guangzhou China

Weakly supervised object localization (WSOL) strives to learn to localize objects with only image-level supervision. Due to the local receptive fields generated by convolution operations, previous CNN-based methods suffer from partial activation issues, concentrating on the object's discriminative part instead of the entire entity scope. Benefiting from the capability of the self-attention mechanism to acquire long-range feature dependencies, Vision Transformer has been recently applied to alleviate the local activation drawbacks. However, since the transformer lacks the inductive localization bias that are inherent in CNNs, it may cause a divergent activation problem resulting in an uncertain distinction between foreground and background. In this work, we proposed a novel Semantic-Constraint Matching Network (SCMN) via transformer to converge on the divergent activation. Specifically, we first propose a local patch shuffle strategy to construct the image pairs, disrupting local patches while guaranteeing global consistency. The paired images that contain the common object in spatial are then fed into the Siamese network encoder. We further design a semantic-constraint matching module, which aims to mine the co-object part by matching the coarse class activation maps (CAMs) extracted from the pair images, thus implicitly guiding and calibrating the transformer network to alleviate the divergent activation. Extensive experimental results conducted on two challenging benchmarks including CUB-200-2011 and ILSVRC datasets show that our method can achieve the new state-of-the-art performance and outperform the previous method by a large margin. Copyright © 2023, The Authors. All rights reserved.

关键词： Chemical activation

New Quantum Algorithms for Computing Quantum Entropies and Distances

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Wang, Qisheng Guan, Ji Liu, Junyi Zhang, Zhicheng Ying, Mingsheng The Graduate School of Mathematics Nagoya University Japan State Key Laboratory of Computer Science Institute of Software Chinese Academy of Sciences China The Joint Center for Quantum Information and Computer Science University of Maryland United States The Centre for Quantum Software and Information University of Technology Sydney Australia

We propose a series of quantum algorithms for computing a wide range of quantum entropies and distances, including the von Neumann entropy, quantum Rényi entropy, trace distance, and fidelity. The proposed algorithms significantly outperform the prior best (and even quantum) ones in the low-rank case, some of which achieve exponential speedups. In particular, for N-dimensional quantum states of rank r, our proposed quantum algorithms for computing the von Neumann entropy, trace distance and fidelity within additive error Ε have time complexity of Õ(r/Ε2), Õ(r5/Ε6) and Õ(r6.5/Ε7.5), respectively. By contrast, prior quantum algorithms for the von Neumann entropy and trace distance usually have time complexity Ω(N), and the prior best one for fidelity has time complexity Õ(r12.5/Ε13.5). The key idea of our quantum algorithms is to extend block-encoding from unitary operators in previous work to quantum states (i.e., density operators). It is realized by developing several convenient techniques to manipulate quantum states and extract information from them. The advantage of our techniques over the existing methods is that no restrictions on density operators are required;in sharp contrast, the previous methods usually require a lower bound on the minimal non-zero eigenvalue of density operators. Copyright © 2022, The Authors. All rights reserved.

关键词： Eigenvalues and eigenfunctions