检索结果-内蒙古大学图书馆

Towards Variable-Length Textual Adversarial Attacks

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Guo, Junliang Zhang, Zhirui Zhang, Linlin Xu, Linli Chen, Boxing Chen, Enhong Luo, Weihua Anhui Province Key Laboratory of Big Data Analysis and Application School of Computer Science and Technology University of Science and Technology of China China Alibaba DAMO Academy Zhejiang University China

Adversarial attacks have shown the vulnerability of machine learning models, however, it is non-trivial to conduct textual adversarial attacks on natural language processing tasks due to the discreteness of data. Most previous approaches conduct attacks with the atomic replacement operation, which usually leads to fixed-length adversarial examples and therefore limits the exploration on the decision space. In this paper, we propose variable-length textual adversarial attacks (VL-Attack) and integrate three atomic operations, namely insertion, deletion and replacement, into a unified framework, by introducing and manipulating a special blank token while attacking. In this way, our approach is able to more comprehensively find adversarial examples around the decision boundary and effectively conduct adversarial attacks. Specifically, our method drops the accuracy of IMDB classification by 96% with only editing 1.3% tokens while attacking a pre-trained BERT model. In addition, fine-tuning the victim model with generated adversarial samples can improve the robustness of the model without hurting the performance, especially for length-sensitive models. On the task of non-autoregressive machine translation, our method can achieve 33.18 BLEU score on IWSLT14 German-English translation, achieving an improvement of 1.47 over the baseline model. Copyright © 2021, The Authors. All rights reserved.

关键词： Natural language processing systems

Spatial fuzzy C-means clustering and deep belief network for change detection in synthetic aperture radar images 5

学校读者我要写书评

暂无评论

Spatial fuzzy C-means clustering and deep belief network for...

IET International Radar Conference 2020, IET IRC 2020

作者： Qi, Wenwen Wu, Lin Guo, Zhengwei Huang, Dan College of Computer and Information Engineering Henan University Kaifeng475004 China Henan Key Laboratory of Big Data Analysis and Processing Henan University Kaifeng475004 China Henan Engineering Research Center of Intelligent Technology and Application Henan University Kaifeng475004 China College of Environment and Planning Henan University Kaifeng475004 China Department of Laboratory and Equipment Management Henan University Kaifeng475004 China

ISBN: (纸本)9781839535635

In this study, spatial fuzzy c-means (SFCM) clustering and deep belief network (DBN) method is presented for change detection in SAR images. There are three primary steps of this approach, they are given as follows: 1) Difference Image (DI) is firstly obtained by calculating the similarity of gray levels between multitemporal SAR images. 2) SFCM clustering is used to analyze DI, which combines the spatial information factor and the fuzzy cluster membership function, thus can effectively eliminate the interference of speckle noise. 3) DBN, a powerful data processing methodology that integrates deep learning techniques to extract underlying features and structures of complicated data, is trained to classify pixels to be changed or unchanged. The effectiveness and robustness of the proposed method are verified by experimental results with the Ottawa dataset. © 2020 Institution of Engineering and Technology. All rights reserved.

关键词： Membership functions

Adaptive normalization for non-stationary time series forecasting: a temporal slice perspective 23

学校读者我要写书评

暂无评论

Adaptive normalization for non-stationary time series foreca...

Proceedings of the 37th International Conference on Neural Information Processing Systems

作者： Zhiding Liu Mingyue Cheng Zhi Li Zhenya Huang Qi Liu Yanhu Xie Enhong Chen Anhui Province Key Laboratory of Big Data Analysis and Application University of Science and Technology of China and State Key Laboratory of Cognitive Intelligence Shenzhen International Graduate School Tsinghua University The First Affiliated Hospital of University of Science and Technology of China

Deep learning models have progressively advanced time series forecasting due to their powerful capacity in capturing sequence dependence. Nevertheless, it is still challenging to make accurate predictions due to the existence of non-stationarity in real-world data, denoting the data distribution rapidly changes over time. To mitigate such a dilemma, several efforts have been conducted by reducing the non-stationarity with normalization operation. However, these methods typically overlook the distribution discrepancy between the input series and the horizon series, and assume that all time points within the same instance share the same statistical properties, which is too ideal and may lead to suboptimal relative improvements. To this end, we propose a novel slice-level adaptive normalization, referred to SAN, which is a novel scheme for empowering time series forecasting with more flexible normalization and denormalization. SAN includes two crucial designs. First, SAN tries to eliminate the non-stationarity of time series in units of a local temporal slice (i.e., sub-series) rather than a global instance. Second, SAN employs a slight network module to independently model the evolving trends of statistical properties of raw time series. Consequently, SAN could serve as a general model-agnostic plugin and better alleviate the impact of the non-stationary nature of time series data. We instantiate the proposed SAN on four widely used forecasting models and test their prediction results on benchmark datasets to evaluate its effectiveness. Also, we report some insightful findings to deeply analyze and understand our proposed SAN. We make our codes publicly available https://***/icantnamemyself/SAN.

关键词：

Modeling Offline Knowledge Evolution Effect for Online Knowledge Tracing

学校读者我要写书评

暂无评论

Modeling Offline Knowledge Evolution Effect for Online Knowl...

International Conference on Tools for Artificial Intelligence (ICTAI)

作者： Xiaonan Zeng Shuanghong Shen Zhenya Huang Enhong Chen Yu Su Shijin Wang Anhui Province Key Laboratory of Big Data Analysis and Application School of Data Science University of Science and Technology of China & State Key Laboratory of Cognitive Intelligence Hefei China Anhui Province Key Laboratory of Big Data Analysis and Application School of Computer Science and Technology University of Science and Technology of China & State Key Laboratory of Cognitive Intelligence Hefei China Hefei Comprehensive National Science Center School of Computer Science and Technology Hefei Normal University & Institute of Artificial Intelligence Hefei China State Key Laboratory of Cognitive Intelligence & iFLYTEK AI Research (Central China) iFLYTEK Co. Ltd Hefei China

Knowledge Tracing (KT) focuses on tracing stu-dents' evolving knowledge state in the process of question-answering, which has been widely applied to online learning systems. Existing KT methods leverage students' online question-answering logs to infer their knowledge state and predict their subsequent performance. However, these methods only focus on students' online behaviors without considering the fact that students' knowledge also evolves offline. Specifically, students have plenty of offline learning actions that also make changes in their knowledge state, which is called the offline knowledge evolution effect in this paper. However, these offline actions are not recorded by online learning systems, therefore leading to unreliable knowledge tracing results. To alleviate this problem, we propose a novel model named offline-aware knowledge tracing (OKT) that considers students' offline knowledge evolution for KT. In O KT, we introduce a novel concept named offline trait to represent student's offline learning action preference. Based on the preference, the model evaluates offline knowledge evolution, and apply the evolution to generate a new knowledge state for students' future performance prediction. To the end, we conduct experiments on three real-world benchmark datasets and demonstrate that OKT achieves state-of-the-art performance. Besides, the experimental results indicate that considering offline actions can help interpret the KT results.

关键词： Learning systems Predictive models Benchmark testing Behavioral sciences

Graph Adaptive Semantic Transfer for Cross-domain Sentiment Classification

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Zhang, Kai Zhang, Kun Liu, Qi Huang, Zhenya Zhang, Mengdi Wu, Wei Cheng, Mingyue Chen, Enhong Anhui Province Key Lab. of Big Data Analysis and Application University of S&T of China State Key Laboratory of Cognitive Intelligence Hefei China School of Computer Science and Information Engineering Hefei University of Technology Hefei China Meituan Beijing China Anhui Province Key Lab. of Big Data Analysis and Application University of S&T of China Hefei China

Cross-domain sentiment classification (CDSC) aims to use the transferable semantics learned from the source domain to predict the sentiment of reviews in the unlabeled target domain. Existing studies in this task attach more attention to the sequence modeling of sentences while largely ignoring the rich domain-invariant semantics embedded in graph structures (i.e., the part-of-speech tags and dependency relations). As an important aspect of exploring characteristics of language comprehension, adaptive graph representations have played an essential role in recent years. To this end, in the paper, we aim to explore the possibility of learning invariant semantic features from graph-like structures in CDSC. Specifically, we present Graph Adaptive Semantic Transfer (GAST) model, an adaptive syntactic graph embedding method that is able to learn domain-invariant semantics from both word sequences and syntactic graphs. More specifically, we first raise a POS-Transformer module to extract sequential semantic features from the word sequences as well as the part-of-speech tags. Then, we design a Hybrid Graph Attention (HGAT) module to generate syntax-based semantic features by considering the transferable dependency relations. Finally, we devise an Integrated aDaptive Strategy (IDS) to guide the joint learning process of both modules. Extensive experiments on four public datasets indicate that GAST achieves comparable effectiveness to a range of state-of-the-art models. © 2022, CC BY.

关键词： Sentiment analysis

A-Map: Interactive Visual Exploration of Intercity Accessibility Dynamics Based on Railway Network data

学校读者我要写书评

暂无评论

A-Map: Interactive Visual Exploration of Intercity Accessibi...

Pacific (formerly Asia-Pacific APVIS) Visualization Symposium

作者： Kaichen Nie Hanning Shao Yuchu Luo Min Tian Hao Wu Wei Zeng Xin Fu Xiaoru Yuan Key Laboratory of Machine Perception (Ministry of Education) School of AI Peking University Hong Kong University of Science and Technology Guangzhou Chang’an University National Engineering Laboratory for Big Data Analysis and Application Peking University

ISBN: (数字)9798350393804

ISBN: (纸本)9798350393811

Railway transportation is closely linked to everyday lives while also aiding domain experts in analyzing national or regional development. However, discrepancies between travel time reduced by railways and actual distance pose challenges in visualizing the accessibility. Existing methods either struggle to simultaneously depict the accessibility relationships among all cities or disregard genuine geographical positions, leading to spatial cognitive confusion. In this work, we propose a novel approach to balance between the geographical positions and travel time between cities. We construct linear cartograms to represent accessibility while preserving better local railway network structures compared with existing methods. We further propose the A-Map system, enabling users to interactively explore the extensive development of China’s railway system over decades. We validate the effectiveness of our proposed layout algorithm from qualitative metric evaluation and illustrate the applicability of our system with results.

关键词： Measurement Visualization Heuristic algorithms Urban areas Layout Transportation data visualization

Stable and Diverse: A Unified Approach for Computerized Adaptive Testing

学校读者我要写书评

暂无评论

Stable and Diverse: A Unified Approach for Computerized Adap...

IEEE International Conference on Cloud Computing and Intelligence Systems (CCIS)

作者： Yuting Ning Ye Liu Zhenya Huang Haoyang Bi Qi Liu Enhong Chen Dan Zhang Anhui Province Key Laboratory of Big Data Analysis and Application School of Computer Science and Technology University of Science and Technology of China School of Data Science University of Science and Technology of China iFLYTEK Research iFLYTEK CO. LTD

ISBN: (纸本)9781665441506

Computerized Adaptive Testing (CAT), aiming to provide personalized tests for each examinee, is an emerging task in the intelligent education field. A CAT system selects questions step by step according to the knowledge states of each examinee, which are estimated by Cognitive Diagnosis Models (CDM). Most existing methods depend on the performance of a single CDM, which is often unstable. Besides, they may select similar questions to generate a test, which to some extent ignores the diversity of selected questions. To this end, in this paper, we propose a novel framework, namely Ensembled Computerized Adaptive Testing (EnCAT). Specifically, EnCAT comprises two components, ensemble part and explore part. In the ensemble part, we ensemble multiple CDMs to determine whether a question is informative, which ensures the stability of CAT process. Then, in the explore part, we learn the question representation from the question content and design a mechanism to quantify the similarity of different questions, which avoids the selection of similar questions and is free of expensive human labeling. Finally, extensive experiments are conducted on a real-world dataset, where the experimental results demonstrate the effectiveness of our proposed EnCAT framework with good performance.

关键词： Adaptation models Cloud computing Computational modeling Conferences Education Stability analysis Labeling

APGL4SR: A Generic Framework with Adaptive and Personalized Global Collaborative Information in Sequential Recommendation

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Yin, Mingjia Wang, Hao Xu, Xiang Wu, Likang Zhao, Sirui Guo, Wei Liu, Yong Tang, Ruiming Lian, Defu Chen, Enhong Anhui Province Key Laboratory of Big Data Analysis and Application University of Science and Technology of China State Key Laboratory of Cognitive Intelligence Hefei China Huawei Singapore Research Center Singapore Huawei Noah's Ark Lab Shenzhen China

The sequential recommendation system has been widely studied for its promising effectiveness in capturing dynamic preferences buried in users' sequential behaviors. Despite the considerable achievements, existing methods usually focus on intra-sequence modeling while overlooking exploiting global collaborative information by inter-sequence modeling, resulting in inferior recommendation performance. Therefore, previous works attempt to tackle this problem with a global collaborative item graph constructed by pre-defined rules. However, these methods neglect two crucial properties when capturing global collaborative information, i.e., adaptiveness and personalization, yielding sub-optimal user representations. To this end, we propose a graph-driven framework, named Adaptive and Personalized Graph Learning for Sequential Recommendation (APGL4SR), that incorporates adaptive and personalized global collaborative information into sequential recommendation systems. Specifically, we first learn an adaptive global graph among all items and capture global collaborative information with it in a self-supervised fashion, whose computational burden can be further alleviated by the proposed SVD-based accelerator. Furthermore, based on the graph, we propose to extract and utilize personalized item correlations in the form of relative positional encoding, which is a highly compatible manner of personalizing the utilization of global collaborative information. Finally, the entire framework is optimized in a multi-task learning paradigm, thus each part of APGL4SR can be mutually reinforced. As a generic framework, APGL4SR can not only outperform other baselines with significant margins, but also exhibit promising versatility, the ability to learn a meaningful global collaborative graph, and the ability to alleviate the dimensional collapse issue of item embeddings. The code is available at https://***/Graph-Team/APGL4SR. © 2023, CC BY.

关键词： Recommender systems

A Novel Approach for Auto-Formulation of Optimization Problems

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Ning, Yuting Liu, Jiayu Qin, Longhu Xiao, Tong Xue, Shangzi Huang, Zhenya Liu, Qi Chen, Enhong Wu, Jinze Anhui Province Key Laboratory of Big Data Analysis and Application School of Computer Science and Technology University of Science and Technology of China China State Key Laboratory of Cognitive Intelligence China iFLYTEK AI Research iFLYTEK Co. Ltd China

In the Natural Language for Optimization (NL4Opt) NeurIPS 2022 competition1, competitors focus on improving the accessibility and usability of optimization solvers, with the aim of subtask 1: recognizing the semantic entities that correspond to the components of the optimization problem;subtask 2: generating formulations for the optimization problem. In this paper, we present the solution of our team. First, we treat subtask 1 as a named entity recognition (NER) problem with the solution pipeline including pre-processing methods, adversarial training, post-processing methods and ensemble learning. Besides, we treat subtask 2 as a generation problem with the solution pipeline including specially designed prompts, adversarial training, post-processing methods and ensemble learning. Our proposed methods have achieved the F1-score of 0.931 in subtask 1 and the accuracy of 0.867 in subtask 2, which won the fourth and third places respectively in this competition. Our code is available at https://***/bigdata-ustc/nl4opt. Copyright © 2023, The Authors. All rights reserved.

关键词： Pipelines