Total: 1343
AI systems that learn through reward feedback about the actions they take are increasingly deployed in domains that have significant impact on our daily life. However, in many cases the online rewards should not be the only guiding criteria, as there are additional constraints and/or priorities imposed by regulations, values, preferences, or ethical principles. We detail a novel online agent that learns a set of behavioral constraints by observation and uses these learned constraints as a guide when making decisions in an online setting while still being reactive to reward feedback. To define this agent, we propose to adopt a novel extension to the classical contextual multi-armed bandit setting and we provide a new algorithm called Behavior Constrained Thompson Sampling (BCTS) that allows for online learning while obeying exogenous constraints. Our agent learns a constrained policy that implements the observed behavioral constraints demonstrated by a teacher agent, and then uses this constrained policy to guide the reward-based online exploration and exploitation. We characterize the upper bound on the expected regret of the contextual bandit algorithm that underlies our agent and provide a case study with real world data in two application domains. Our experiments show that the designed agent is able to act within the set of behavior constraints without significantly degrading its overall reward performance.
Attributed network embedding has received much interest from the research community as most of the networks come with some content in each node, which is also known as node attributes. Existing attributed network approaches work well when the network is consistent in structure and attributes, and nodes behave as expected. But real world networks often have anomalous nodes. Typically these outliers, being relatively unexplainable, affect the embeddings of other nodes in the network. Thus all the downstream network mining tasks fail miserably in the presence of such outliers. Hence an integrated approach to detect anomalies and reduce their overall effect on the network embedding is required. Towards this end, we propose an unsupervised outlier aware network embedding algorithm (ONE) for attributed networks, which minimizes the effect of the outlier nodes, and hence generates robust network embeddings. We align and jointly optimize the loss functions coming from structure and attributes of the network. To the best of our knowledge, this is the first generic network embedding approach which incorporates the effect of outliers for an attributed network without any supervision. We experimented on publicly available real networks and manually planted different types of outliers to check the performance of the proposed algorithm. Results demonstrate the superiority of our approach to detect the network outliers compared to the state-of-the-art approaches. We also consider different downstream machine learning applications on networks to show the efficiency of ONE as a generic network embedding technique. The source code is made available at https://github.com/sambaranban/ONE.
Thispaperconsidersextractivesummarisationinacomparative setting: given two or more document groups (e.g., separated by publication time), the goal is to select a small number of documents that are representative of each group, and also maximally distinguishable from other groups. We formulate a set of new objective functions for this problem that connect recent literature on document summarisation, interpretable machine learning, and data subset selection. In particular, by casting the problem as a binary classification amongst different groups, we derive objectives based on the notion of maximum mean discrepancy, as well as a simple yet effective gradient-based optimisation strategy. Our new formulation allows scalable evaluations of comparative summarisation as a classification task, both automatically and via crowd-sourcing. To this end, we evaluate comparative summarisation methods on a newly curated collection of controversial news topics over 13months.Weobserve thatgradient-based optimisationoutperforms discrete and baseline approaches in 15 out of 24 different automatic evaluation settings. In crowd-sourced evaluations, summaries from gradient optimisation elicit 7% more accurate classification from human workers than discrete optimisation. Our result contrasts with recent literature on submodular data subset selection that favours discrete optimisation. We posit that our formulation of comparative summarisation will prove useful in a diverse range of use cases such as comparing content sources, authors, related topics, or distinct view points.
Automatically annotating column types with knowledge base (KB) concepts is a critical task to gain a basic understanding of web tables. Current methods rely on either table metadata like column name or entity correspondences of cells in the KB, and may fail to deal with growing web tables with incomplete meta information. In this paper we propose a neural network based column type annotation framework named ColNet which is able to integrate KB reasoning and lookup with machine learning and can automatically train Convolutional Neural Networks for prediction. The prediction model not only considers the contextual semantics within a cell using word representation, but also embeds the semantics of a column by learning locality features from multiple cells. The method is evaluated with DBPedia and two different web table datasets, T2Dv2 from the general Web and Limaye from Wikipedia pages, and achieves higher performance than the state-of-the-art approaches.
One-class collaborative filtering (OCCF) problems are vital in many applications of recommender systems, such as news and music recommendation, but suffers from sparsity issues and lacks negative examples. To address this problem, the state-of-the-arts assigned smaller weights to unobserved samples and performed low-rank approximation. However, the ground-truth ratings of unobserved samples are usually set to zero but ill-defined. In this paper, we propose a ranking-based implicit regularizer and provide a new general framework for OCCF, to avert the ground-truth ratings of unobserved samples. We then exploit it to regularize a ranking-based loss function and design efficient optimization algorithms to learn model parameters. Finally, we evaluate them on three realworld datasets. The results show that the proposed regularizer significantly improves ranking-based algorithms and that the proposed framework outperforms the state-of-the-art OCCF algorithms.
Online Shopping has become a part of our daily routine, but it still cannot offer intuitive experience as store shopping. Nowadays, most e-commerce Websites offer a Question Answering (QA) system that allows users to consult other users who have purchased the product. However, users still need to wait patiently for others’ replies. In this paper, we investigate how to provide a quick response to the asker by plausible answer identification from product reviews. By analyzing the similarity and discrepancy between explicit answers and reviews that can be answers, a novel multi-task deep learning method with carefully designed attention mechanisms is developed. The method can well exploit large amounts of user generated QA data and a few manually labeled review data to address the problem. Experiments on data collected from Amazon demonstrate its effectiveness and superiority over competitive baselines.
Providing explanations in a recommender system is getting more and more attention in both industry and research communities. Most existing explainable recommender models regard user preferences as invariant to generate static explanations. However, in real scenarios, a user’s preference is always dynamic, and she may be interested in different product features at different states. The mismatching between the explanation and user preference may degrade costumers’ satisfaction, confidence and trust for the recommender system. With the desire to fill up this gap, in this paper, we build a novel Dynamic Explainable Recommender (called DER) for more accurate user modeling and explanations. In specific, we design a time-aware gated recurrent unit (GRU) to model user dynamic preferences, and profile an item by its review information based on sentence-level convolutional neural network (CNN). By attentively learning the important review information according to the user current state, we are not only able to improve the recommendation performance, but also can provide explanations tailored for the users’ current preferences. We conduct extensive experiments to demonstrate the superiority of our model for improving recommendation performance. And to evaluate the explainability of our model, we first present examples to provide intuitive analysis on the highlighted review information, and then crowd-sourcing based evaluations are conducted to quantitatively verify our model’s superiority.
In general, recommendation can be viewed as a matching problem, i.e., match proper items for proper users. However, due to the huge semantic gap between users and items, it’s almost impossible to directly match users and items in their initial representation spaces. To solve this problem, many methods have been studied, which can be generally categorized into two types, i.e., representation learning-based CF methods and matching function learning-based CF methods. Representation learning-based CF methods try to map users and items into a common representation space. In this case, the higher similarity between a user and an item in that space implies they match better. Matching function learning-based CF methods try to directly learn the complex matching function that maps user-item pairs to matching scores. Although both methods are well developed, they suffer from two fundamental flaws, i.e., the limited expressiveness of dot product and the weakness in capturing low-rank relations respectively. To this end, we propose a general framework named DeepCF, short for Deep Collaborative Filtering, to combine the strengths of the two types of methods and overcome such flaws. Extensive experiments on four publicly available datasets demonstrate the effectiveness of the proposed DeepCF framework.
Spreadsheet table detection is the task of detecting all tables on a given sheet and locating their respective ranges. Automatic table detection is a key enabling technique and an initial step in spreadsheet data intelligence. However, the detection task is challenged by the diversity of table structures and table layouts on the spreadsheet. Considering the analogy between a cell matrix as spreadsheet and a pixel matrix as image, and encouraged by the successful application of Convolutional Neural Networks (CNN) in computer vision, we have developed TableSense, a novel end-to-end framework for spreadsheet table detection. First, we devise an effective cell featurization scheme to better leverage the rich information in each cell; second, we develop an enhanced convolutional neural network model for table detection to meet the domain-specific requirement on precise table boundary detection; third, we propose an effective uncertainty metric to guide an active learning based smart sampling algorithm, which enables the efficient build-up of a training dataset with 22,176 tables on 10,220 sheets with broad coverage of diverse table structures and layouts. Our evaluation shows that TableSense is highly effective with 91.3% recall and 86.5% precision in EoB-2 metric, a significant improvement over both the current detection algorithm that are used in commodity spreadsheet tools and state-of-the-art convolutional neural networks in computer vision.
A Triple in knowledge-graph takes a form that consists of head, relation, tail. Triple Classification is used to determine the truth value of an unknown Triple. This is a hard task for 1-to-N relations using the vector-based embedding approach. We propose a new region-based embedding approach using fine-grained type chains. A novel geometric process is presented to extend the vectors of pre-trained entities into n-balls (n-dimensional balls) under the condition that head balls shall contain their tail balls. Our algorithm achieves zero energy cost, therefore, serves as a case study of perfectly imposing tree structures into vector space. An unknown Triple (h,r,x) will be predicted as true, when x’s n-ball is located in the r-subspace of h’s n-ball, following the same construction of known tails of h. The experiments are based on large datasets derived from the benchmark datasets WN11, FB13, and WN18. Our results show that the performance of the new method is related to the length of the type chain and the quality of pre-trained entityembeddings, and that performances of long chains with welltrained entity-embeddings outperform other methods in the literature. Source codes and datasets are located at https: //github.com/GnodIsNait/mushroom.
With the promising progress of deep neural networks, layer aggregation has been used to fuse information across layers in various fields, such as computer vision and machine translation. However, most of the previous methods combine layers in a static fashion in that their aggregation strategy is independent of specific hidden states. Inspired by recent progress on capsule networks, in this paper we propose to use routing-by-agreement strategies to aggregate layers dynamically. Specifically, the algorithm learns the probability of a part (individual layer representations) assigned to a whole (aggregated representations) in an iterative way and combines parts accordingly. We implement our algorithm on top of the state-of-the-art neural machine translation model TRANSFORMER and conduct experiments on the widely-used WMT14 sh⇒German and WMT17 Chinese⇒English translation datasets. Experimental results across language pairs show that the proposed approach consistently outperforms the strong baseline model and a representative static aggregation model.
As one promising way to solve the challenging issues of data sparsity and cold start in recommender systems, crossdomain recommendation has gained increasing research interest recently. Cross-domain recommendation aims to improve the recommendation performance by means of transferring explicit or implicit feedback from the auxiliary domain to the target domain. Although the side information of review texts and item contents has been proven to be useful in recommendation, most existing works only use one kind of side information and cannot deeply fuse this side information with ratings. In this paper, we propose a Review and Content based Deep Fusion Model named RC-DFM for crossdomain recommendation. We first extend Stacked Denoising Autoencoders (SDAE) to effectively fuse review texts and item contents with the rating matrix in both auxiliary and target domains. Through this way, the learned latent factors of users and items in both domains preserve more semantic information for recommendation. Then we utilize a multi-layer perceptron to transfer user latent factors between the two domains to address the data sparsity and cold start issues. Experimental results on real datasets demonstrate the superior performance of RC-DFM compared with state-of-the-art recommendation methods. Deeply Fusing Reviews and Contents for Cold Start Users in Cross-Domain Recommendation Systems
Real web datasets are often associated with multiple views such as long and short commentaries, users preference and so on. However, with the rapid growth of user generated texts, each view of the dataset has a large feature space and leads to the computational challenge during matrix decomposition process. In this paper, we propose a novel multi-view clustering algorithm based on the non-negative matrix factorization that attempts to use feature sampling strategy in order to reduce the complexity during the iteration process. In particular, our method exploits unsupervised semantic information in the learning process to capture the intrinsic similarity through a graph regularization. Moreover, we use Hilbert Schmidt Independence Criterion (HSIC) to explore the unsupervised semantic diversity information among multi-view contents of one web item. The overall objective is to minimize the loss function of multi-view non-negative matrix factorization that combines with an intra-semantic similarity graph regularizer and an inter-semantic diversity term. Compared with some state-of-the-art methods, we demonstrate the effectiveness of our proposed method on a large real-world dataset Doucom and the other three smaller datasets.
The advent of social media has presented a promising new opportunity for the early detection of depression. To do so effectively, there are two challenges to overcome. The first is that textual and visual information must be jointly considered to make accurate inferences about depression. The second challenge is that due to the variety of content types posted by users, it is difficult to extract many of the relevant indicator texts and images. In this work, we propose the use of a novel cooperative multi-agent model to address these challenges. From the historical posts of users, the proposed method can automatically select related indicator texts and images. Experimental results demonstrate that the proposed method outperforms state-of-the-art methods by a large margin (over 30% error reduction). In several experiments and examples, we also verify that the selected posts can successfully indicate user depression, and our model can obtained a robust performance in realistic scenarios.
Clustering on multi-view data has attracted much more attention in the past decades. Most previous studies assume that each instance appears in all views, or there is at least one view containing all instances. However, real world data often suffers from missing some instances in each view, leading to the research problem of partial multi-view clustering. To address this issue, this paper proposes a simple yet effective Anchorbased Partial Multi-view Clustering (APMC) method, which utilizes anchors to reconstruct instance-to-instance relationships for clustering. APMC is conceptually simple and easy to implement in practice, besides it has clear intuitions and non-trivial empirical guarantees. Specifically, APMC firstly integrates intra- and inter- view similarities through anchors. Then, spectral clustering is performed on the fused similarities to obtain a unified clustering result. Compared with existing partial multi-view clustering methods, APMC has three notable advantages: 1) it can capture more non-linear relations among instances with the help of kernel-based similarities; 2) it has a much lower time complexity in virtue of a noniterative scheme; 3) it can inherently handle data with negative entries as well as be extended to more than two views. Finally, we extensively evaluate the proposed method on five benchmark datasets. Experimental results demonstrate the superiority of APMC over state-of-the-art approaches.
Jointly learning representations of 3D shapes and text is crucial to support tasks such as cross-modal retrieval or shape captioning. A recent method employs 3D voxels to represent 3D shapes, but this limits the approach to low resolutions due to the computational cost caused by the cubic complexity of 3D voxels. Hence the method suffers from a lack of detailed geometry. To resolve this issue, we propose Y2Seq2Seq, a view-based model, to learn cross-modal representations by joint reconstruction and prediction of view and word sequences. Specifically, the network architecture of Y2Seq2Seq bridges the semantic meaning embedded in the two modalities by two coupled “Y” like sequence-tosequence (Seq2Seq) structures. In addition, our novel hierarchical constraints further increase the discriminability of the cross-modal representations by employing more detailed discriminative information. Experimental results on cross-modal retrieval and 3D shape captioning show that Y2Seq2Seq outperforms the state-of-the-art methods.
Customers ask questions, and customer service staffs answer those questions. It is the basic service manner of customer service (CS). The progress of CS is a typical multi-round conversation. However, there are no explicit corresponding relations among conversational utterances. This paper focuses on obtaining explicit alignments of question and answer utterances in CS. It not only is an important task of dialogue analysis, but also able to obtain lots of valuable train data for learning dialogue systems. In this work, we propose end-to-end models for aligning question (Q) and answer (A) utterances in CS conversation with recurrent pointer networks (RPN). On the one hand, RPN-based alignment models are able to model the conversational contexts and the mutual influence of different Q-A alignments. On the other hand, they are able to address the issue of empty and multiple alignments for some utterances in a unified manner. We construct a dataset from an in-house online CS. The experimental results demonstrate that the proposed models are effective to learn the alignments of question and answer utterances.
This paper proposes a novel method for generating compact answers to open-domain why-questions, such as the following answer, “Because deep learning technologies were introduced,” to the question, “Why did Google’s machine translation service improve so drastically?” Although many works have dealt with why-question answering, most have focused on retrieving as answers relatively long text passages that consist of several sentences. Because of their length, such passages are not appropriate to be read aloud by spoken dialog systems and smart speakers; hence, we need to create a method that generates compact answers. We developed a novel neural summarizer for this compact answer generation task. It combines a recurrent neural network-based encoderdecoder model with stacked convolutional neural networks and was designed to effectively exploit background knowledge, in this case a set of causal relations (e.g., “[Microsoft’s machine translation has made great progress over the last few years]effect since [it started to use deep learning.]cause”) that was extracted from a large web data archive (4 billion web pages). Our experimental results show that our method achieved significantly better ROUGE F-scores than existing encoder-decoder models and their variations that were augmented with query-attention and memory networks, which are used to exploit the background knowledge.
Community detection is a fundamental problem in network science with various applications. The problem has attracted much attention and many approaches have been proposed. Among the existing approaches are the latest methods based on Graph Convolutional Networks (GCN) and on statistical modeling of Markov Random Fields (MRF). Here, we propose to integrate the techniques of GCN and MRF to solve the problem of semi-supervised community detection in attributed networks with semantic information. Our new method takes advantage of salient features of GNN and MRF and exploits both network topology and node semantic information in a complete end-to-end deep network architecture. Our extensive experiments demonstrate the superior performance of the new method over state-of-the-art methods and its scalability on several large benchmark problems.
Recent research on community detection focuses on learning representations of nodes using different network embedding methods, and then feeding them as normal features to clustering algorithms. However, we find that though one may have good results by direct clustering based on such network embedding features, there is ample room for improvement. More seriously, in many real networks, some statisticallysignificant nodes which play pivotal roles are often divided into incorrect communities using network embedding methods. This is because while some distance measures are used to capture the spatial relationship between nodes by embedding, the nodes after mapping to feature vectors are essentially not coupled any more, losing important structural information. To address this problem, we propose a general Markov Random Field (MRF) framework to incorporate coupling in network embedding which allows better detecting network communities. By smartly utilizing properties of MRF, the new framework not only preserves the advantages of network embedding (e.g. low complexity, high parallelizability and applicability for traditional machine learning), but also alleviates its core drawback of inadequate representations of dependencies via making up the missing coupling relationships. Experiments on real networks show that our new approach improves the accuracy of existing embedding methods (e.g. Node2Vec, DeepWalk and MNMF), and corrects most wrongly-divided statistically-significant nodes, which makes network embedding essentially suitable for real community detection applications. The new approach also outperforms other state-of-the-art conventional community detection methods.
We examine the problem of crawling the community structure of a multiplex network containing multiple layers of edge relationships. While there has been a great deal of work examining community structure in general, and some work on the problem of sampling a network to preserve its community structure, to the best of our knowledge, this is the first work to consider this problem on multiplex networks. We consider the specific case in which the layers of a multiplex network have different query (collection) costs and reliabilities; and a data collector is interested in identifying the community structure of the most expensive layer. We propose MultiComSample (MCS), a novel algorithm for crawling a multiplex network. MCS uses multiple levels of multi-armed bandits to determine the best layers, communities and node roles for selecting nodes to query. We test MCS against six baseline algorithms on real-world multiplex networks, and achieved large gains in performance. For example, after consuming a budget equivalent to sampling 20% of the nodes in the expensive layer, we observe that MCS outperforms the best baseline by up to 49%.
In recent years, hashing has attracted more and more attention owing to its superior capacity of low storage cost and high query efficiency in large-scale cross-modal retrieval. Benefiting from deep leaning, continuously compelling results in cross-modal retrieval community have been achieved. However, existing deep cross-modal hashing methods either rely on amounts of labeled information or have no ability to learn an accuracy correlation between different modalities. In this paper, we proposed Unsupervised coupled Cycle generative adversarial Hashing networks (UCH), for cross-modal retrieval, where outer-cycle network is used to learn powerful common representation, and inner-cycle network is explained to generate reliable hash codes. Specifically, our proposed UCH seamlessly couples these two networks with generative adversarial mechanism, which can be optimized simultaneously to learn representation and hash codes. Extensive experiments on three popular benchmark datasets show that the proposed UCH outperforms the state-of-the-art unsupervised cross-modal hashing methods.
The task of user ranking in signed networks, aiming to predict potential friends and enemies for each user, has attracted increasing attention in numerous applications. Existing approaches are mainly extended from heuristics of the traditional models in unsigned networks. They suffer from two limitations: (1) mainly focus on global rankings thus cannot provide effective personalized ranking results, and (2) have a relatively unrealistic assumption that each user treats her neighbors’ social strengths indifferently. To address these two issues, we propose a supervised method based on random walk to learn social strengths between each user and her neighbors, in which the random walk more likely visits “potential friends” and less likely visits “potential enemies”. We learn the personalized social strengths by optimizing on a particularly designed loss function oriented on ranking. We further present a fast ranking method based on the local structure among each seed node and a certain set of candidates. It much simplifies the proposed ranking model meanwhile maintains the performance. Experimental results demonstrate the superiority of our approach over the state-of-the-art approaches.
Question Routing (QR) on Community-based Question Answering (CQA) websites aims at recommending answerers that have high probabilities of providing the “accepted answers” to new questions. The existing question routing algorithms simply predict the ranking of users based on query content. As a consequence, the question raiser information is ignored. On the other hand, they lack learnable scoring functions to explicitly compute ranking scores. To tackle these challenges, we propose NeRank that (1) jointly learns representations of question content, question raiser, and question answerers by a heterogeneous information network embedding algorithm and a long short-term memory (LSTM) model. The embeddings of the three types of entities are unified in the same latent space, and (2) conducts question routing for personalized queries, i.e., queries with two entities (question content, question raiser), by a convolutional scoring function taking the learned embeddings of all three types of entities as input. Using the scores, NeRank routes new questions to high-ranking answerers that are skillfulness in the question domain and have similar backgrounds to the question raiser. Experimental results show that NeRank significantly outperforms competitive baseline question routing models that ignore the raiser information in three ranking metrics. In addition, NeRank is convergeable in several thousand iterations and insensitive to parameter changes, which prove its effectiveness, scalability, and robustness.
Predicting the popularity of online article sheds light to many applications such as recommendation, advertising and information retrieval. However, there are several technical challenges to be addressed for developing the best of predictive capability. (1) The popularity fluctuates under impacts of external factors, which are unpredictable and hard to capture. (2) Content and meta-data features, largely determining the online content popularity, are usually multi-modal and nontrivial to model. (3) Besides, it also needs to figure out how to integrate temporal process and content features modeling for popularity prediction in different lifecycle stages of online articles. In this paper, we propose a Deep Fusion of Temporal process and Content features (DFTC) method to tackle them. For modeling the temporal popularity process, we adopt the recurrent neural network and convolutional neural network. For multi-modal content features, we exploit the hierarchical attention network and embedding technique. Finally, a temporal attention fusion is employed for dynamically integrating all these parts. Using datasets collected from WeChat, we show that the proposed model significantly outperforms state-of-the-art approaches on popularity prediction.