Open Access. Powered by Scholars. Published by Universities.®

Digital Commons Network™

Open Access. Powered by Scholars. Published by Universities.®

Research Collection School Of Computing and Information Systems

Discipline
Keyword
Publication Year
File Type

Articles 3001 - 3030 of 8597

Full-Text Articles in Entire DC Network

Privacy-Preserving Federated Deep Learning With Irregular Users, Guowen Xu, Hongwei Li, Yun Zhang, Shengmin Xu, Jianting Ning, Robert H. Deng Mar 2021

Privacy-Preserving Federated Deep Learning With Irregular Users, Guowen Xu, Hongwei Li, Yun Zhang, Shengmin Xu, Jianting Ning, Robert H. Deng

Research Collection School Of Computing and Information Systems

Federated deep learning has been widely used in various fields. To protect data privacy, many privacy-preserving approaches have also been designed and implemented in various scenarios. However, existing works rarely consider a fundamental issue that the data shared by certain users (called irregular users) may be of low quality. Obviously, in a federated training process, data shared by many irregular users may impair the training accuracy, or worse, lead to the uselessness of the final model. In this paper, we propose PPFDL, a Privacy-Preserving Federated Deep Learning framework with irregular users. In specific, we design a novel solution to reduce …

Go to article

Investigating The Adoption Of Hybrid Encrypted Cloud Data Deduplication With Game Theory, Xueqin Liang, Zheng Yan, Robert H. Deng, Qinghu Zheng Mar 2021

Investigating The Adoption Of Hybrid Encrypted Cloud Data Deduplication With Game Theory, Xueqin Liang, Zheng Yan, Robert H. Deng, Qinghu Zheng

Research Collection School Of Computing and Information Systems

Encrypted data deduplication, along with different preferences in data access control, brings the birth of hybrid encrypted cloud data deduplication (H-DEDU for short). However, whether H-DEDU can be successfully deployed in practice has not been seriously investigated. Obviously, the adoption of H-DEDU depends on whether it can bring economic benefits to all stakeholders. But existing economic models of cloud storage fail to support H-DEDU due to complicated interactions among stakeholders. In this article, we establish a formal economic model of H-DEDU by formulating the utilities of all involved stakeholders, i.e., data holders, data owners, and Cloud Storage Providers (CSPs). Then, …

Go to article

Combining Query Reduction And Expansion For Text-Retrieval-Based Bug Localization, Juan Manuel Florez, Oscar Chaparro, Christoph Treude, Andrian Marcus Mar 2021

Combining Query Reduction And Expansion For Text-Retrieval-Based Bug Localization, Juan Manuel Florez, Oscar Chaparro, Christoph Treude, Andrian Marcus

Research Collection School Of Computing and Information Systems

Automated text-retrieval-based bug localization (TRBL) techniques normally use the full text of a bug report to formulate a query and retrieve parts of the code that are buggy. Previous research has shown that reducing the size of the query increases the effectiveness of TRBL. On the other hand, researchers also found improvements when expanding the query (i.e., adding more terms). In this paper, we bring these two views together to reformulate queries for TRBL. Specifically, we improve discourse-based query reduction strategies, by adopting a combinatorial approach and using task phrases from bug reports, and combine them with a state-of-the-art query …

Go to article

Deep Learning For Anomaly Detection: Challenges, Methods, And Opportunities, Guansong Pang, Longbing Cao, Charu Aggarwal Mar 2021

Deep Learning For Anomaly Detection: Challenges, Methods, And Opportunities, Guansong Pang, Longbing Cao, Charu Aggarwal

Research Collection School Of Computing and Information Systems

In this tutorial we aim to present a comprehensive survey of the advances in deep learning techniques specifically designed for anomaly detection (deep anomaly detection for short). Deep learning has gained tremendous success in transforming many data mining and machine learning tasks, but popular deep learning techniques are inapplicable to anomaly detection due to some unique characteristics of anomalies, e.g., rarity, heterogeneity, boundless nature, and prohibitively high cost of collecting large-scale anomaly data. Through this tutorial, audiences would gain a systematic overview of this area, learn the key intuitions, objective functions, underlying assumptions, advantages and disadvantages of different categories of …

Go to article

Enhancing Healthcare Professional And Caregiving Staff Informedness With Data Analytics For Chronic Disease Management, Na Liu, Robert John Kauffman Mar 2021

Enhancing Healthcare Professional And Caregiving Staff Informedness With Data Analytics For Chronic Disease Management, Na Liu, Robert John Kauffman

Research Collection School Of Computing and Information Systems

An important area in healthcare to which data analytics can be applied is chronic disease management. The chronic care model is mostly patient-centric, so patients have been considered as the end users of data analytics. The information needs of healthcare providers have been overlooked. Drawing upon the theory of informedness and the transtheoretical model of health behavior change, we use a multicase study approach to investigate the information needs of different caregiving stakeholders in the spectrum of chronic diseases, and how data analytics can be designed to meet the varying needs of professionals and staff to support their informedness.

Go to article

Is The Ground Truth Really Accurate? Dataset Purification For Automated Program Repair, Deheng Yang, Yan Lei, Xiaoguang Mao, David Lo, Huan Xie, Meng Yan Mar 2021

Is The Ground Truth Really Accurate? Dataset Purification For Automated Program Repair, Deheng Yang, Yan Lei, Xiaoguang Mao, David Lo, Huan Xie, Meng Yan

Research Collection School Of Computing and Information Systems

Datasets of real-world bugs shipped with human-written patches are intensively used in the evaluation of existing automated program repair (APR) techniques, wherein the human-written patches always serve as the ground truth, for manual or automated assessment approaches, to evaluate the correctness of test-suite adequate patches. An inaccurate human-written patch tangled with other code changes will pose threats to the reliability of the assessment results. Therefore, the construction of such datasets always requires much manual effort on isolating real bug fixes from bug fixing commits. However, the manual work is time-consuming and prone to mistakes, and little has been known on …

Go to article

Bilateral Variational Autoencoder For Collaborative Filtering, Quoc Tuan Truong, Aghiles Salah, Hady W. Lauw Mar 2021

Bilateral Variational Autoencoder For Collaborative Filtering, Quoc Tuan Truong, Aghiles Salah, Hady W. Lauw

Research Collection School Of Computing and Information Systems

Preference data is a form of dyadic data, with measurements associated with pairs of elements arising from two discrete sets of objects. These are users and items, as well as their interactions, e.g., ratings. We are interested in learning representations for both sets of objects, i.e., users and items, to predict unknown pairwise interactions. Motivated by the recent successes of deep latent variable models, we propose Bilateral Variational Autoencoder (BiVAE), which arises from a combination of a generative model of dyadic data with two inference models, user- and item-based, parameterized by neural networks. Interestingly, our model can take the form …

Go to article

Fast Scene Labeling Via Structural Inference, Huaidong Zhang, Chu Han, Xiaodan Zhang, Yong Du, Xuemiao Xu, Guoqiang Han, Jing Qin, Shengfeng He Mar 2021

Fast Scene Labeling Via Structural Inference, Huaidong Zhang, Chu Han, Xiaodan Zhang, Yong Du, Xuemiao Xu, Guoqiang Han, Jing Qin, Shengfeng He

Research Collection School Of Computing and Information Systems

Scene labeling or parsing aims to assign pixelwise semantic labels for an input image. Existing CNN-based models cannot leverage the label dependencies, while RNN-based models predict labels within the local context. In this paper, we propose a fast LSTM scene labeling network via structural inference. A minimum spanning tree is used to build the image structure for constructing semantic relationships. This structure allows efficient generation of direct parent-child dependencies for arbitrary levels of superpixels, and thus structural relationships can be learned with LSTM. In particular, we propose a bi-directional recurrent network to model the information flow along the parent-child path. …

Go to article

How Do Monetary Incentives Influence Giving? An Empirical Investigation Of Matching Subsidies On Kiva, Zhiyuan Gao, Zhiling Guo, Qian Tang Mar 2021

How Do Monetary Incentives Influence Giving? An Empirical Investigation Of Matching Subsidies On Kiva, Zhiyuan Gao, Zhiling Guo, Qian Tang

Research Collection School Of Computing and Information Systems

Matching subsidies, through which third-party institutions provide a dollar-for-dollar match of private contributions made through selected campaigns, have served as effective tools to boost fundraising. We utilize a quasi-experiment on a prosocial crowdfunding platform to examine the effectiveness of matching subsidies in shaping funding outcomes and lender behaviors. Although matching subsidies offer matched loans competitive advantages over unmatched loans, we find that total private contributions made to both matched and unmatched loans increase compared to their prematching counterparts, suggesting a positive spillover effect on unmatched loans. However, matching subsidies lead to decreased private contributions made on the platform after a …

Go to article

Learning To Assess The Quality Of Stroke Rehabilitation Exercises, Min Hun Lee, Daniel P. Siewiorek, Asim Smailagic, Alexandre Bernardino, Sergi Bermúdez I Badia Mar 2021

Learning To Assess The Quality Of Stroke Rehabilitation Exercises, Min Hun Lee, Daniel P. Siewiorek, Asim Smailagic, Alexandre Bernardino, Sergi Bermúdez I Badia

Research Collection School Of Computing and Information Systems

Due to the limited number of therapists, task-oriented exercises are often prescribed for post-stroke survivors as in-home rehabilitation. During in-home rehabilitation, a patient may become unmotivated or confused to comply prescriptions without the feedback of a therapist. To address this challenge, this paper proposes an automated method that can achieve not only qualitative, but also quantitative assessment of stroke rehabilitation exercises. Specifically, we explored a threshold model that utilizes the outputs of binary classifiers to quantify the correctness of a movements into a performance score. We collected movements of 11 healthy subjects and 15 post-stroke survivors using a Kinect sensor …

Go to article

Privacy-Preserving Multi-Keyword Searchable Encryption For Distributed Systems, Xueqiao Liu, Guomin Yang, Willy Susilo, Joseph Tonien, Jian Shen Mar 2021

Privacy-Preserving Multi-Keyword Searchable Encryption For Distributed Systems, Xueqiao Liu, Guomin Yang, Willy Susilo, Joseph Tonien, Jian Shen

Research Collection School Of Computing and Information Systems

As cloud storage has been widely adopted in various applications, how to protect data privacy while allowing efficient data search and retrieval in a distributed environment remains a challenging research problem. Existing searchable encryption schemes are still inadequate on desired functionality and security/privacy perspectives. Specifically, supporting multi-keyword search under the multi-user setting, hiding search pattern and access pattern, and resisting keyword guessing attacks (KGA) are the most challenging tasks. In this article, we present a new searchable encryption scheme that addresses the above problems simultaneously, which makes it practical to be adopted in distributed systems. It not only enables multi-keyword …

Go to article

Traceable Monero: Anonymous Cryptocurrency With Enhanced Accountability, Yannan Li, Guomin Yang, Wily Susilo, Yong Yu, Man Ho Au, Dongxi Liu Mar 2021

Traceable Monero: Anonymous Cryptocurrency With Enhanced Accountability, Yannan Li, Guomin Yang, Wily Susilo, Yong Yu, Man Ho Au, Dongxi Liu

Research Collection School Of Computing and Information Systems

Monero provides a high level of anonymity for both users and their transactions. However, many criminal activities might be committed with the protection of anonymity in cryptocurrency transactions. Thus, user accountability (or traceability) is also important in Monero transactions, which is unfortunately lacking in the current literature. In this paper, we fill this gap by introducing a new cryptocurrency named Traceable Monero to balance the user anonymity and accountability. Our framework relies on a tracing authority, but is optimistic, in that it is only involved when investigations in certain transactions are required. We formalize the system model and security model …

Go to article

Can We Classify Cashless Payment Solution Implementations At The Country Level?, Dennis Ng, Robert J. Kauffman, Paul Robert Griffin Mar 2021

Can We Classify Cashless Payment Solution Implementations At The Country Level?, Dennis Ng, Robert J. Kauffman, Paul Robert Griffin

Research Collection School Of Computing and Information Systems

This research commentary proposes a 3-D implementation classification framework to assist service providers and business leaders in understanding the kinds of contexts in which more or less successful cashless payment solutions are observed at point-of-sale (PoS) settings. Three constructs characterize the framework: the digitalization of the local implementation environment; the relative novelty of a given payment technology solution in a country at a specific point in time; and the development status of the country’s national infrastructure. The framework is motivated by a need to support cross-country research in this domain. We analyze eight country mini-cases based on an eight-facet (2 …

Go to article

Improving Multi-Hop Knowledge Base Question Answering By Learning Intermediate Supervision Signals, Gaole He, Yunshi Lan, Jing Jiang, Wayne Xin Zhao, Ji Rong Wen Mar 2021

Improving Multi-Hop Knowledge Base Question Answering By Learning Intermediate Supervision Signals, Gaole He, Yunshi Lan, Jing Jiang, Wayne Xin Zhao, Ji Rong Wen

Research Collection School Of Computing and Information Systems

Multi-hop Knowledge Base Question Answering (KBQA) aims to find the answer entities that are multiple hops away in the Knowledge Base (KB) from the entities in the question. A major challenge is the lack of supervision signals at intermediate steps. Therefore, multi-hop KBQA algorithms can only receive the feedback from the final answer, which makes the learning unstable or ineffective. To address this challenge, we propose a novel teacher-student approach for the multi-hop KBQA task. In our approach, the student network aims to find the correct answer to the query, while the teacher network tries to learn intermediate supervision signals …

Go to article

All The Wiser: Fake News Intervention Using User Reading Preferences, Kuan Chieh Lo, Shih Chieh Dai, Aiping Xiong, Jing Jiang, Lun Wei Ku Mar 2021

All The Wiser: Fake News Intervention Using User Reading Preferences, Kuan Chieh Lo, Shih Chieh Dai, Aiping Xiong, Jing Jiang, Lun Wei Ku

Research Collection School Of Computing and Information Systems

To address the increasingly significant issue of fake news, we develop a news reading platform in which we propose an implicit approach to reduce people's belief in fake news. Specifically, we leverage reinforcement learning to learn an intervention module on top of a recommender system (RS) such that the module is activated to replace RS to recommend news toward the verification once users touch the fake news. To examine the effect of the proposed method, we conduct a comprehensive evaluation with 89 human subjects and check the effective rate of change in belief but without their other limitations. Moreover, 84% …

Go to article

Assessing Code Clone Harmfulness: Indicators, Factors, And Counter Measures, Bin Hu, Yijian Wu, Xin Peng, Jun Sun, Nanjie Zhan, Jun Wu Mar 2021

Assessing Code Clone Harmfulness: Indicators, Factors, And Counter Measures, Bin Hu, Yijian Wu, Xin Peng, Jun Sun, Nanjie Zhan, Jun Wu

Research Collection School Of Computing and Information Systems

Code clones are identical or similar code in software projects. On one hand, developers clone code to achieve higher productivity and thus clones inherently exist; on the other hand, code clones demand extra effort to maintain the consistency between clone instances and may introduce bugs, and thus are often considered harmful for software maintenance and quality. We believe that not all code clones have the same level of harmfulness. A systematic way of assessing the harmfulness level of cloned code would facilitate informed decisions on how to deal with clones. We propose a model for clone harmfulness level assessment with …

Go to article

Evoking Empathy: A Framework For Describing Empathy Tools, Sydney Pratte, Anthony Tang, Lora Oehlberg Feb 2021

Evoking Empathy: A Framework For Describing Empathy Tools, Sydney Pratte, Anthony Tang, Lora Oehlberg

Research Collection School Of Computing and Information Systems

Empathy tools are experiences designed to evoke empathetic responses by placing the user in another’s lived and felt experience. The problem is that designers do not have a common vocabulary to describe empathy tool experiences; consequently, it is difficult to compare/contrast empathy tool designs or to think about their efficacy. To address this problem, we analyzed 26 publications on empathy tools to develop a descriptive framework for designers of empathy tools. Based on our analysis, we found that empathy tools can be described along three dimensions: (i) the amount of agency the tool allows, (ii) the user’s perspective while using …

Go to article

Mimoa: A Membrane-Inspired Multi-Objective Algorithm For Green Vehicle Routing Problem With Stochastic Demands, Yunyun Niu, Yongpeng Zhang, Zhiguang Cao, Kaizhou Gao, Jianhua Xiao, Wen Song, Fangwei Zhang Feb 2021

Mimoa: A Membrane-Inspired Multi-Objective Algorithm For Green Vehicle Routing Problem With Stochastic Demands, Yunyun Niu, Yongpeng Zhang, Zhiguang Cao, Kaizhou Gao, Jianhua Xiao, Wen Song, Fangwei Zhang

Research Collection School Of Computing and Information Systems

Nowadays, an increasing number of vehicle routing problem with stochastic demands (VRPSD) models have been studied to meet realistic needs in the field of logistics. In this paper, a bi-objective vehicle routing problem with stochastic demands (BO-VRPSD) was investigated, which aims to minimize total cost and customer dissatisfaction. Different from traditional vehicle routing problem (VRP) models, both the uncertainty in customer demands and the nature of multiple objectives make the problem more challenging. To cope with BO-VRPSD, a membrane-inspired multi-objective algorithm (MIMOA) was proposed, which is characterized by a parallel distributed framework with two operation subsystems and one control subsystem, …

Go to article

Adversarial Meta Sampling For Multilingual Low-Resource Speech Recognition, Yubei Xiao, Ke Gong, Pan Zhou, Guolin Zheng, Xiaodan Liang, Liang Lin Feb 2021

Adversarial Meta Sampling For Multilingual Low-Resource Speech Recognition, Yubei Xiao, Ke Gong, Pan Zhou, Guolin Zheng, Xiaodan Liang, Liang Lin

Research Collection School Of Computing and Information Systems

Human doctors with well-structured medical knowledge can diagnose a disease merely via a few conversations with patients about symptoms. In contrast, existing knowledgegrounded dialogue systems often require a large number of dialogue instances to learn as they fail to capture the correlations between different diseases and neglect the diagnostic experience shared among them. To address this issue, we propose a more natural and practical paradigm, i.e., low-resource medical dialogue generation, which can transfer the diagnostic experience from source diseases to target ones with a handful of data for adaptation. It is capitalized on a commonsense knowledge graph to characterize the …

Go to article

Divide And Capture: An Improved Cryptanalysis Of The Encryption Standard Algorithm Rsa, Willy Susilo, Joseph Tonien, Guomin Yang Feb 2021

Divide And Capture: An Improved Cryptanalysis Of The Encryption Standard Algorithm Rsa, Willy Susilo, Joseph Tonien, Guomin Yang

Research Collection School Of Computing and Information Systems

RSA is a well known standard algorithm used by modern computers to encrypt and decrypt messages. In some applications, to save the decryption time, it is desirable to have a short secret key d compared to the modulus N. The first significant attack that breaks RSA with short secret key given by Wiener in 1990 is based on the continued fraction technique and it works with d < 1/4 root 18 N-.(25). A decade later, in 2000, Boneh and Durfee presented an improved attack based on lattice technique which works with d < N-.(292). Until this day, Boneh-Durfee attack remain as the best attack on RSA with short secret key. In this paper, we revisit the continued fraction technique and propose a new attack on RSA. Our main result shows that when d < root t (2 root 2 + 8/3) N-.(75)/root e, where e is the public exponent and t is a chosen parameter, our attack can break the RSA with the running time of O(tlog (N)). Our attack is especially well suited for the case where e is much smaller than N. When e approximate to N, the Boneh-Durfee attack outperforms ours. As a result, we could simultaneously run both attacks, our new attack and the classical Boneh-Durfee attack as a backup.

Go to article

Fault Analysis And Debugging Of Microservice Systems: Industrial Survey, Benchmark System, And Empirical Study, Xiang Zhou, Xin Peng, Tao Xie, Jun Sun, Chao Ji, Wenhai Li, Dan Ding Feb 2021

Fault Analysis And Debugging Of Microservice Systems: Industrial Survey, Benchmark System, And Empirical Study, Xiang Zhou, Xin Peng, Tao Xie, Jun Sun, Chao Ji, Wenhai Li, Dan Ding

Research Collection School Of Computing and Information Systems

The complexity and dynamism of microservice systems pose unique challenges to a variety of software engineering tasks such as fault analysis and debugging. In spite of the prevalence and importance of microservices in industry, there is limited research on the fault analysis and debugging of microservice systems. To fill this gap, we conduct an industrial survey to learn typical faults of microservice systems, current practice of debugging, and the challenges faced by developers in practice. We then develop a medium-size benchmark microservice system (being the largest and most complex open source microservice system within our knowledge) and replicate 22 industrial …

Go to article

Graph-Evolving Meta-Learning For Low-Resource Medical Dialogue Generation, Shuai Lin, Pan Zhou, Xiaodan Liang, Jianheng Tang, Ruihui Zhao, Ziliang Chen, Liang Lin Feb 2021

Graph-Evolving Meta-Learning For Low-Resource Medical Dialogue Generation, Shuai Lin, Pan Zhou, Xiaodan Liang, Jianheng Tang, Ruihui Zhao, Ziliang Chen, Liang Lin

Research Collection School Of Computing and Information Systems

Human doctors with well-structured medical knowledge can diagnose a disease merely via a few conversations with patients about symptoms. In contrast, existing knowledge-grounded dialogue systems often require a large number of dialogue instances to learn as they fail to capture the correlations between different diseases and neglect the diagnostic experience shared among them. To address this issue, we propose a more natural and practical paradigm, i.e., low-resource medical dialogue generation, which can transfer the diagnostic experience from source diseases to target ones with a handful of data for adaptation. It is capitalized on a commonsense knowledge graph to characterize the …

Go to article

Scalable Verification Of Quantized Neural Networks, Thomas A. Henzinger, Mathias Lechner, Dorde Zikelic Feb 2021

Scalable Verification Of Quantized Neural Networks, Thomas A. Henzinger, Mathias Lechner, Dorde Zikelic

Research Collection School Of Computing and Information Systems

Formal verification of neural networks is an active topic of research, and recent advances have significantly increased the size of the networks that verification tools can handle. However, most methods are designed for verification of an idealized model of the actual network which works over real arithmetic and ignores rounding imprecisions. This idealization is in stark contrast to network quantization, which is a technique that trades numerical precision for computational efficiency and is, therefore, often applied in practice. Neglecting rounding errors of such low-bit quantized neural networks has been shown to lead to wrong conclusions about the network’s correctness. Thus, …

Go to article

An Exploratory Study On The Introduction And Removal Of Different Types Of Technical Debt In Deep Learning Frameworks, Jiakun Liu, Qiao Huang, Xin Xia, Emad Shihab, David Lo, Shanping Li Feb 2021

An Exploratory Study On The Introduction And Removal Of Different Types Of Technical Debt In Deep Learning Frameworks, Jiakun Liu, Qiao Huang, Xin Xia, Emad Shihab, David Lo, Shanping Li

Research Collection School Of Computing and Information Systems

To complete tasks faster, developers often have to sacrifice the quality of the software. Such compromised practice results in the increasing burden to developers in future development. The metaphor, technical debt, describes such practice. Prior research has illustrated the negative impact of technical debt, and many researchers investigated how developers deal with a certain type of technical debt. However, few studies focused on the removal of different types of technical debt in practice. To fill this gap, we use the introduction and removal of different types of self-admitted technical debt (i.e., SATD) in 7 deep learning frameworks as an example. …

Go to article

Understanding Adversarial Robustness Via Critical Attacking Route, Tianlin Li, Aishan Liu, Xianglong Liu, Yitao Xu, Chongzhi Zhang, Xiaofei Xie Feb 2021

Understanding Adversarial Robustness Via Critical Attacking Route, Tianlin Li, Aishan Liu, Xianglong Liu, Yitao Xu, Chongzhi Zhang, Xiaofei Xie

Research Collection School Of Computing and Information Systems

Deep neural networks (DNNs) are vulnerable to adversarial examples which are generated by inputs with imperceptible perturbations. Understanding adversarial robustness of DNNs has become an important issue, which would for certain result in better practical deep learning applications. To address this issue, we try to explain adversarial robustness for deep models from a new perspective of critical attacking route, which is computed by a gradient-based influence propagation strategy. Similar to rumor spreading in social net-works, we believe that adversarial noises are amplified and propagated through the critical attacking route. By exploiting neurons' influences layer by layer, we compose the critical …

Go to article

Norm-Based Generalisation Bounds For Deep Multi-Class Convolutional Neural Networks, Antoine Ledent, Waleed Mustafa, Yunwen Lei, Marius Kloft Feb 2021

Norm-Based Generalisation Bounds For Deep Multi-Class Convolutional Neural Networks, Antoine Ledent, Waleed Mustafa, Yunwen Lei, Marius Kloft

Research Collection School Of Computing and Information Systems

We show generalisation error bounds for deep learning with two main improvements over the state of the art. (1) Our bounds have no explicit dependence on the number of classes except for logarithmic factors. This holds even when formulating the bounds in terms of the Frobenius-norm of the weight matrices, where previous bounds exhibit at least a squareroot dependence on the number of classes. (2) We adapt the classic Rademacher analysis of DNNs to incorporate weight sharing—a task of fundamental theoretical importance which was previously attempted only under very restrictive assumptions. In our results, each convolutional filter contributes only once …

Go to article

Fine-Grained Generalization Analysis Of Vector-Valued Learning, Liang Wu, Antoine Ledent, Yunwen Lei, Marius Kloft Feb 2021

Fine-Grained Generalization Analysis Of Vector-Valued Learning, Liang Wu, Antoine Ledent, Yunwen Lei, Marius Kloft

Research Collection School Of Computing and Information Systems

Many fundamental machine learning tasks can be formulated as a problem of learning with vector-valued functions, where we learn multiple scalar-valued functions together. Although there is some generalization analysis on different specific algorithms under the empirical risk minimization principle, a unifying analysis of vector-valued learning under a regularization framework is still lacking. In this paper, we initiate the generalization analysis of regularized vector-valued learning algorithms by presenting bounds with a mild dependency on the output dimension and a fast rate on the sample size. Our discussions relax the existing assumptions on the restrictive constraint of hypothesis spaces, smoothness of loss …

Go to article

Neural Architecture Search As Sparse Supernet, Y. Wu, A. Liu, Zhiwu Huang, S. Zhang, Gool L. Van Feb 2021

Neural Architecture Search As Sparse Supernet, Y. Wu, A. Liu, Zhiwu Huang, S. Zhang, Gool L. Van

Research Collection School Of Computing and Information Systems

This paper aims at enlarging the problem of Neural Architecture Search (NAS) from Single-Path and Multi-Path Search to automated Mixed-Path Search. In particular, we model the NAS problem as a sparse supernet using a new continuous architecture representation with a mixture of sparsity constraints. The sparse supernet enables us to automatically achieve sparsely-mixed paths upon a compact set of nodes. To optimize the proposed sparse supernet, we exploit a hierarchical accelerated proximal gradient algorithm within a bi-level optimization framework. Extensive experiments on Convolutional Neural Network and Recurrent Neural Network search demonstrate that the proposed method is capable of searching for …

Go to article

Treecaps: Tree-Based Capsule Networks For Source Code Processing, Duy Quoc Nghi Bui, Yijun Yu, Lingxiao Jiang Feb 2021

Treecaps: Tree-Based Capsule Networks For Source Code Processing, Duy Quoc Nghi Bui, Yijun Yu, Lingxiao Jiang

Research Collection School Of Computing and Information Systems

Recently program learning techniques have been proposed to process source code based on syntactical structures (e.g., Abstract Syntax Trees) and/or semantic information (e.g., Dependency Graphs). While graphs may be better at capturing various viewpoints of code semantics than trees, constructing graph inputs from code need static code semantic analysis that may not be accurate and introduces noise during learning. On the other hand, syntax trees are precisely defined according to the language grammar and easier to construct and process than graphs. We propose a new tree-based learning technique, named TreeCaps, by fusing capsule networks with tree-based convolutional neural networks, to …

Go to article

Multi-Decoder Attention Model With Embedding Glimpse For Solving Vehicle Routing Problems, Liang Xin, Wen Song, Zhiguang Cao, Jie Zhang Feb 2021

Multi-Decoder Attention Model With Embedding Glimpse For Solving Vehicle Routing Problems, Liang Xin, Wen Song, Zhiguang Cao, Jie Zhang

Research Collection School Of Computing and Information Systems

We present a novel deep reinforcement learning method to learn construction heuristics for vehicle routing problems. In specific, we propose a Multi-Decoder Attention Model (MDAM) to train multiple diverse policies, which effectively increases the chance of finding good solutions compared with existing methods that train only one policy. A customized beam search strategy is designed to fully exploit the diversity of MDAM. In addition, we propose an Embedding Glimpse layer in MDAM based on the recursive nature of construction, which can improve the quality of each policy by providing more informative embeddings. Extensive experiments on six different routing problems show …

Go to article

First Prev Next Last