检索结果-内蒙古大学图书馆

A retrieval-based dialogue system utilizing utterance and context embeddings

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Bartl, Alexander Spanakis, Gerasimos Department of Data Science and Knowledge Engineering Maastricht University MD6200 Netherlands

Finding semantically rich and computerunderstandable representations for textual dialogues, utterances and words is crucial for dialogue systems (or conversational agents), as their performance mostly depends on understanding the context of conversations. Recent research aims at finding distributed vector representations (embeddings) for words, such that semantically similar words are relatively close within the vector-space. Encoding the "meaning" of text into vectors is a current trend, and text can range from words, phrases and documents to actual human-to-human conversations. In recent research approaches, responses have been generated utilizing a decoder architecture, given the vector representation of the current conversation. In this paper, the utilization of embeddings for answer retrieval is explored by using Locality-Sensitive Hashing Forest (LSH Forest), an Approximate Nearest Neighbor (ANN) model, to find similar conversations in a corpus and rank possible candidates. Experimental results on the well-known Ubuntu Corpus (in English) and a customer service chat dataset (in Dutch) show that, in combination with a candidate selection method, retrieval-based approaches outperform generative ones and reveal promising future research directions towards the usability of such a system. Copyright © 2017, The Authors. All rights reserved.

关键词： Nearest neighbor search

Dual encoding for zero-example video retrieval

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Dong, Jianfeng Li, Xirong Xu, Chaoxi Ji, Shouling He, Yuan Yang, Gang Wang, Xun College of Computer and Information Engineering Zhejiang Gongshang University Key Lab of Data Engineering and Knowledge Engineering Renmin University of China AI & Media Computing Lab School of Information Renmin University of China College of Computer Science Zhejiang University Alibaba-Zhejiang University Joint Research Institute of Frontier Technologies Alibaba Group

This paper attacks the challenging problem of zero-example video retrieval. In such a retrieval paradigm, an end user searches for unlabeled videos by ad-hoc queries described in natural language text with no visual example provided. Given videos as sequences of frames and queries as sequences of words, an effective sequence-to-sequence cross-modal matching is required. The majority of existing methods are concept based, extracting relevant concepts from queries and videos and accordingly establishing associations between the two modalities. In contrast, this paper takes a concept-free approach, proposing a dual deep encoding network that encodes videos and queries into powerful dense representations of their own. Dual encoding is conceptually simple, practically effective and end-to-end. As experiments on three benchmarks, i.e. MSR-VTT, TRECVID 2016 and 2017 Ad-hoc Video Search show, the proposed solution establishes a new state-of-the-art for zero-example video retrieval. Copyright © 2018, The Authors. All rights reserved.

关键词： Signal encoding

A generalized concept-cognitive learning: A machine learning viewpoint

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Mi, Yunlong Shi, Yong Li, Jinhai School of Computer and Control Engineering University of Chinese Academy of Sciences Beijing101408 China Key Laboratory of Big Data Mining and Knowledge Management Chinese Academy of Sciences Beijing100190 China Research Center on Fictitious Economy and Data Science Chinese Academy of Sciences Beijing100190 China College of Information Science and Technology University of Nebraska at Omaha NE68182 United States Faculty of Science Kunming University of Science and Technology Kunming650500 China

Concept-cognitive learning (CCL) is a hot topic in recent years, and it has attracted much attention from the communities of formal concept analysis, granular computing and cognitive computing. However, the relationship among cognitive computing (CC), concept-cognitive computing (CCC), CCL and concept-cognitive learning model (CCLM) is not clearly described. To this end, we first explain the relationship of CC, CCC, CCL and CCLM. Then, we propose a generalized concept-cognitive learning (GCCL) from the point of view of machine learning. Finally, experiments on some data sets are conducted to verify the feasibility of concept formation and concept-cognitive process of GCCL. Copyright © 2018, The Authors. All rights reserved.

关键词： Machine learning

Exploring anti-community structure in networks with application to incompatibility of traditional Chinese medicine

学校读者我要写书评

暂无评论

PHYSICA A-STATISTICAL MECHANICS AND ITS APPLICATIONS 2017年 486卷 31-43页

作者： Zhu, Jiajing Liu, Yongguo Zhang, Yun Liu, Xiaofeng Xiao, Yonghua Wang, Shidong Wu, Xindong [a]Knowledge and Data Engineering Laboratory of Chinese Medicine School of Information and Software Engineering University of Electronic Science and Technology of China Chengdu 610054 PR China [b]School of Basic Medical Science Beijing University of Chinese Medicine Beijing 100029 PR China [c]Dongzhimen Hospital Beijing University of Chinese Medicine Beijing 100700 PR China [d]School of Computing and Informatics University of Louisiana at Lafayette Lafayette LA 70503 USA

Community structure is one of the most important properties in networks, in which a node shares its most connections with the others in the same community. On the contrary, the anti-community structure means the nodes in the same group have few or no connections with each other. In Traditional Chinese Medicine (TCM), the incompatibility problem of herbs is a challenge to the clinical medication safety. In this paper, we propose a new anti community detection algorithm, Random non-nEighboring nOde expansioN (REON), to find anti-communities in networks, in which a new evaluation criterion, anti-modularity, is designed to measure the quality of the obtained anti-community structure. In order to establish anti-communities in REON, we expand the node set by non-neighboring node expansion and regard the node set with the highest anti-modularity as an anti-community. Inspired by the phenomenon that the node with higher degree has greater contribution to the anti-modularity, an improved algorithm called REONI is developed by expanding node set by the non-neighboring node with the maximum degree, which greatly enhances the efficiency of REON. Experiments on synthetic and real-world networks demonstrate the superiority of the proposed algorithms over the existing methods. In addition, by applying REONI to the herb network, we find that it can discover incompatible herb combinations. (C) 2017 Elsevier B.V. All rights reserved.

关键词： Complex networks Anti-community detection Traditional Chinese medicine Incompatibility of herbs

Online Robust Lagrangian Support Vector Machine against Adversarial Attack

学校读者我要写书评

暂无评论

Procedia Computer science 2018年 139卷 173-181页

作者： Yue Ma Yiwei He Yingjie Tian School of Mathematical Sciences University of Chinese Academy of Sciences Beijing 100049 China Research Center on Fictitious Economy and Data Science Chinese Academy of Sciences Beijing 100190 China School of Computer and Control Engineering University of Chinese Academy of Sciences Beijing 100049 China Key Laboratory of Big Data Mining and Knowledge management Beijing 100190 China School of Economics and Management University of Chinese Academy of Sciences Beijing 100190 China

In adversarial environment such as intrusion detection and spam filtering, the adversary-intruder or spam advertiser may attempt to produce contaminate training instance and manipulate the learning of classifier. In order to keep good classification performance, many robuster learning methods have been proposed to deal with the adversarial attack. Support Vector Machines(SVMs) is a kind of successful approach in the adversarial classification tasks and the investigation of robust SVMs is very popular. However, in many real application, the data including stain instance is coming dynamically. Batch learning which needs retraining when encountering new samples, will consume more computing resources. In this paper, we propose a robust Lagrangian support vector machine (RLSVM) with modified kernel matrix and explore the online learning algorithm on it. The experimental results show the robustness of RLSVM against label noise produced by adversaries under the online adversarial environment.

关键词： adversarial attack poison attack label noise online learning Lagrangian SVM

Massive open online courses temporal profiling for dropout prediction

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Hagedoorn, Tom Rolandus Spanakis, Gerasimos Department of Data Science and Knowledge Engineering Maastricht University 6200MD Netherlands

Massive Open Online Courses (MOOCs) are attracting the attention of people all over the world. Regardless the platform, numbers of registrants for online courses are impressive but in the same time, completion rates are disappointing. Understanding the mechanisms of dropping out based on the learner profile arises as a crucial task in MOOCs, since it will allow intervening at the right moment in order to assist the learner in completing the course. In this paper, the dropout behaviour of learners in a MOOC is thoroughly studied by first extracting features that describe the behavior of learners within the course and then by comparing three classifiers (Logistic Regression, Random Forest and AdaBoost) in two tasks: Predicting which users will have dropped out by a certain week and predicting which users will drop out on a specific week. The former has showed to be considerably easier, with all three classifiers performing equally well. However, the accuracy for the second task is lower, and Logistic Regression tends to perform slightly better than the other two algorithms. We found that features that reflect an active attitude of the user towards the MOOC, such as submitting their assignment, posting on the Forum and filling their Profile, are strong indicators of persistence. Copyright © 2017, The Authors. All rights reserved.

关键词： Forecasting

Effects of synaptic and myelin plasticity on learning in a network of kuramoto phase oscillators

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Karimian, M. Dibenedetto, D. Moerel, M. Burwick, T. Westra, R.L. De Weerd, P. Senden, M. Maastricht University Maastricht6229 ER Netherlands Department of Cognitive Neuroscience Faculty of Psychology and Neuroscience Maastricht University P.O. Box 616 Maastricht6200 MD Netherlands Maastricht Brain Imaging Centre Faculty of Psychology and Neuroscience Maastricht University P.O. Box 616 Maastricht6200 MD Netherlands Department of Data Science and Knowledge Engineering Maastricht University Maastricht6211 LH Netherlands Frankfurt Institute for Advanced Studies Goethe University Frankfurt Ruth-Moufang-Str. 1 Frankfurt am Main60438 Germany

Models of learning typically focus on synaptic plasticity. However, learning is the result of both synaptic and myelin plasticity. Specifically, synaptic changes often co-occur and interact with myelin changes, leading to complex dynamic interactions between these processes. Here, we investigate the implications of these interactions for the coupling behavior of a system of Kuramoto oscillators. To that end, we construct a fully connected, one-dimensional ring network of phase oscillators whose coupling strength (reflecting synaptic strength) as well as conduction velocity (reflecting myelination) are each regulated by a Hebbian learning rule. We evaluate the behavior of the system in terms of structural (pairwise connection strength and conduction velocity) and functional connectivity (local and global synchronization behavior). We find that for conditions in which a system limited to synaptic plasticity develops two distinct clusters both structurally and functionally, additional adaptive myelination allows for functional communication across these structural clusters. Hence, dynamic conduction velocity permits the functional integration of structurally segregated clusters. Our results confirm that network states following learning may be different when myelin plasticity is considered in addition to synaptic plasticity, pointing towards the relevance of integrating both factors in computational models of *** Codes 34C15 Copyright © 2019, The Authors. All rights reserved.

关键词： Synchronization

On the combination of coevolution and novelty search

学校读者我要写书评

暂无评论

On the combination of coevolution and novelty search

Congress on Evolutionary Computation

作者： Fabian Fränz Jan Paredis Rico Möckel Department of Data Science and Knowledge Engineering Maastricht University Maastricht The Netherlands Universiteit Maastricht Maastricht Limburg NL

This paper develops a new method for coevolution, named Fitness-Diversity Driven Coevolution (FDDC). This approach builds on existing methods by a combination of a (predator-prey) Coevolutionary Genetic Algorithm (CGA) and novelty search. The innovation lies in replacing the absolute novelty measure with a relative one, called Fitness-Diversity. FDDC overcomes problems common in both CGAs (premature convergence and unbalanced coevolution) and in novelty search (construction of an archive). As a proof of principle, Spring Loaded Inverted Pendulums (SLIPs) are coevolved with 2D-terrains the SLIPs must learn to traverse.

关键词： Sociology Statistics Springs Linear programming Genetic algorithms Optimization Search problems

It’s Time to Consider ‘Time’ when Evaluating Recommender-System Algorithms [Proposal]

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Beel, Joeran Trinity College Dublin Department of Computer Science and Statistics Knowledge and Data Engineering Group ADAPT Centre Ireland

In this position paper, we question the current practice of calculating evaluation metrics for recommender systems as single numbers (e.g. precision p=.28 or mean absolute error MAE = 1.21). We argue that single numbers express only average effectiveness over usually a rather long period (e.g. a year or even longer). This practice provides only a vague and static view of the data. We propose that recommender-system researchers should instead calculate metrics for time-series such as weeks or months, and plot the results in e.g. a line chart. This way, results show how algorithms’ effectiveness develops over time, and hence the results allow drawing more meaningful conclusions about how an algorithm will perform in the future. In this paper, we explain our reasoning, provide an example to illustrate our reasoning and present suggestions for what the community should do next. Copyright © 2017, The Authors. All rights reserved.

关键词： Recommender systems