July 28, 2019

2907 words 14 mins read

Paper Group ANR 257

Towards Large-Pose Face Frontalization in the Wild. Modeling Semantic Expectation: Using Script Knowledge for Referent Prediction. Higher-Order Minimum Cost Lifted Multicuts for Motion Segmentation. Predicting Rich Drug-Drug Interactions via Biomedical Knowledge Graphs and Text Jointly Embedding. Persistent-homology-based gait recognition. Using RD …

Towards Large-Pose Face Frontalization in the Wild


Title	Towards Large-Pose Face Frontalization in the Wild
Authors	Xi Yin, Xiang Yu, Kihyuk Sohn, Xiaoming Liu, Manmohan Chandraker
Abstract	Despite recent advances in face recognition using deep learning, severe accuracy drops are observed for large pose variations in unconstrained environments. Learning pose-invariant features is one solution, but needs expensively labeled large-scale data and carefully designed feature learning algorithms. In this work, we focus on frontalizing faces in the wild under various head poses, including extreme profile views. We propose a novel deep 3D Morphable Model (3DMM) conditioned Face Frontalization Generative Adversarial Network (GAN), termed as FF-GAN, to generate neutral head pose face images. Our framework differs from both traditional GANs and 3DMM based modeling. Incorporating 3DMM into the GAN structure provides shape and appearance priors for fast convergence with less training data, while also supporting end-to-end training. The 3DMM-conditioned GAN employs not only the discriminator and generator loss but also a new masked symmetry loss to retain visual quality under occlusions, besides an identity loss to recover high frequency information. Experiments on face recognition, landmark localization and 3D reconstruction consistently show the advantage of our frontalization method on faces in the wild datasets.
Tasks	3D Reconstruction, Face Recognition
Published	2017-04-20
URL	http://arxiv.org/abs/1704.06244v3
PDF	http://arxiv.org/pdf/1704.06244v3.pdf
PWC	https://paperswithcode.com/paper/towards-large-pose-face-frontalization-in-the
Repo
Framework

Modeling Semantic Expectation: Using Script Knowledge for Referent Prediction


Title	Modeling Semantic Expectation: Using Script Knowledge for Referent Prediction
Authors	Ashutosh Modi, Ivan Titov, Vera Demberg, Asad Sayeed, Manfred Pinkal
Abstract	Recent research in psycholinguistics has provided increasing evidence that humans predict upcoming content. Prediction also affects perception and might be a key to robustness in human language processing. In this paper, we investigate the factors that affect human prediction by building a computational model that can predict upcoming discourse referents based on linguistic knowledge alone vs. linguistic knowledge jointly with common-sense knowledge in the form of scripts. We find that script knowledge significantly improves model estimates of human predictions. In a second study, we test the highly controversial hypothesis that predictability influences referring expression type but do not find evidence for such an effect.
Tasks	Common Sense Reasoning
Published	2017-02-10
URL	http://arxiv.org/abs/1702.03121v1
PDF	http://arxiv.org/pdf/1702.03121v1.pdf
PWC	https://paperswithcode.com/paper/modeling-semantic-expectation-using-script
Repo
Framework

Higher-Order Minimum Cost Lifted Multicuts for Motion Segmentation


Title	Higher-Order Minimum Cost Lifted Multicuts for Motion Segmentation
Authors	Margret Keuper
Abstract	Most state-of-the-art motion segmentation algorithms draw their potential from modeling motion differences of local entities such as point trajectories in terms of pairwise potentials in graphical models. Inference in instances of minimum cost multicut problems defined on such graphs al- lows to optimize the number of the resulting segments along with the segment assignment. However, pairwise potentials limit the discriminative power of the employed motion models to translational differences. More complex models such as Euclidean or affine transformations call for higher-order potentials and a tractable inference in the resulting higher-order graphical models. In this paper, we (1) introduce a generalization of the minimum cost lifted multicut problem to hypergraphs, and (2) propose a simple primal feasible heuristic that allows for a reasonably efficient inference in instances of higher-order lifted multicut problem instances defined on point trajectory hypergraphs for motion segmentation. The resulting motion segmentations improve over the state-of-the-art on the FBMS-59 dataset.
Tasks	Motion Segmentation
Published	2017-04-06
URL	http://arxiv.org/abs/1704.01811v2
PDF	http://arxiv.org/pdf/1704.01811v2.pdf
PWC	https://paperswithcode.com/paper/higher-order-minimum-cost-lifted-multicuts
Repo
Framework

Predicting Rich Drug-Drug Interactions via Biomedical Knowledge Graphs and Text Jointly Embedding


Title	Predicting Rich Drug-Drug Interactions via Biomedical Knowledge Graphs and Text Jointly Embedding
Authors	Meng Wang
Abstract	Minimizing adverse reactions caused by drug-drug interactions has always been a momentous research topic in clinical pharmacology. Detecting all possible interactions through clinical studies before a drug is released to the market is a demanding task. The power of big data is opening up new approaches to discover various drug-drug interactions. However, these discoveries contain a huge amount of noise and provide knowledge bases far from complete and trustworthy ones to be utilized. Most existing studies focus on predicting binary drug-drug interactions between drug pairs but ignore other interactions. In this paper, we propose a novel framework, called PRD, to predict drug-drug interactions. The framework uses the graph embedding that can overcome data incompleteness and sparsity issues to achieve multiple DDI label prediction. First, a large-scale drug knowledge graph is generated from different sources. Then, the knowledge graph is embedded with comprehensive biomedical text into a common low dimensional space. Finally, the learned embeddings are used to efficiently compute rich DDI information through a link prediction process. To validate the effectiveness of the proposed framework, extensive experiments were conducted on real-world datasets. The results demonstrate that our model outperforms several state-of-the-art baseline methods in terms of capability and accuracy.
Tasks	Graph Embedding, Knowledge Graphs, Link Prediction
Published	2017-12-24
URL	http://arxiv.org/abs/1712.08875v4
PDF	http://arxiv.org/pdf/1712.08875v4.pdf
PWC	https://paperswithcode.com/paper/predicting-rich-drug-drug-interactions-via
Repo
Framework

Persistent-homology-based gait recognition


Title	Persistent-homology-based gait recognition
Authors	J. Lamar-Leon, Raul Alonso-Baryolo, Edel Garcia-Reyes, R. Gonzalez-Diaz
Abstract	Gait recognition is an important biometric technique for video surveillance tasks, due to the advantage of using it at distance. In this paper, we present a persistent homology-based method to extract topological features (the so-called {\it topological gait signature}) from the the body silhouettes of a gait sequence. % It has been used before in several conference papers of the same authors for human identification, gender classification, carried object detection and monitoring human activities at distance. % The novelty of this paper is the study of the stability of the topological gait signature under small perturbations and the number of gait cycles contained in a gait sequence. In other words, we show that the topological gait signature is robust to the presence of noise in the body silhouettes and to the number of gait cycles contained in a given gait sequence. % We also show that computing our topological gait signature of only the lowest fourth part of the body silhouette, we avoid the upper body movements that are unrelated to the natural dynamic of the gait, caused for example by carrying a bag or wearing a coat.
Tasks	Gait Recognition, Object Detection
Published	2017-07-21
URL	http://arxiv.org/abs/1707.06982v1
PDF	http://arxiv.org/pdf/1707.06982v1.pdf
PWC	https://paperswithcode.com/paper/persistent-homology-based-gait-recognition
Repo
Framework

Using RDF Summary Graph For Keyword-based Semantic Searches


Title	Using RDF Summary Graph For Keyword-based Semantic Searches
Authors	Serkan Ayvaz, Mehmet Aydar
Abstract	The Semantic Web began to emerge as its standards and technologies developed rapidly in the recent years. The continuing development of Semantic Web technologies has facilitated publishing explicit semantics with data on the Web in RDF data model. This study proposes a semantic search framework to support efficient keyword-based semantic search on RDF data utilizing near neighbor explorations. The framework augments the search results with the resources in close proximity by utilizing the entity type semantics. Along with the search results, the system generates a relevance confidence score measuring the inferred semantic relatedness of returned entities based on the degree of similarity. Furthermore, the evaluations assessing the effectiveness of the framework and the accuracy of the results are presented.
Tasks
Published	2017-07-12
URL	http://arxiv.org/abs/1707.03602v1
PDF	http://arxiv.org/pdf/1707.03602v1.pdf
PWC	https://paperswithcode.com/paper/using-rdf-summary-graph-for-keyword-based
Repo
Framework

Elephant Search with Deep Learning for Microarray Data Analysis


Title	Elephant Search with Deep Learning for Microarray Data Analysis
Authors	Mrutyunjaya Panda
Abstract	Even though there is a plethora of research in Microarray gene expression data analysis, still, it poses challenges for researchers to effectively and efficiently analyze the large yet complex expression of genes. The feature (gene) selection method is of paramount importance for understanding the differences in biological and non-biological variation between samples. In order to address this problem, a novel elephant search (ES) based optimization is proposed to select best gene expressions from the large volume of microarray data. Further, a promising machine learning method is envisioned to leverage such high dimensional and complex microarray dataset for extracting hidden patterns inside to make a meaningful prediction and most accurate classification. In particular, stochastic gradient descent based Deep learning (DL) with softmax activation function is then used on the reduced features (genes) for better classification of different samples according to their gene expression levels. The experiments are carried out on nine most popular Cancer microarray gene selection datasets, obtained from UCI machine learning repository. The empirical results obtained by the proposed elephant search based deep learning (ESDL) approach are compared with most recent published article for its suitability in future Bioinformatics research.
Tasks
Published	2017-07-12
URL	http://arxiv.org/abs/1707.03604v1
PDF	http://arxiv.org/pdf/1707.03604v1.pdf
PWC	https://paperswithcode.com/paper/elephant-search-with-deep-learning-for
Repo
Framework

Real-Time Energy Disaggregation of a Distribution Feeder’s Demand Using Online Learning


Title	Real-Time Energy Disaggregation of a Distribution Feeder’s Demand Using Online Learning
Authors	Gregory S. Ledva, Laura Balzano, Johanna L. Mathieu
Abstract	Though distribution system operators have been adding more sensors to their networks, they still often lack an accurate real-time picture of the behavior of distributed energy resources such as demand responsive electric loads and residential solar generation. Such information could improve system reliability, economic efficiency, and environmental impact. Rather than installing additional, costly sensing and communication infrastructure to obtain additional real-time information, it may be possible to use existing sensing capabilities and leverage knowledge about the system to reduce the need for new infrastructure. In this paper, we disaggregate a distribution feeder’s demand measurements into: 1) the demand of a population of air conditioners, and 2) the demand of the remaining loads connected to the feeder. We use an online learning algorithm, Dynamic Fixed Share (DFS), that uses the real-time distribution feeder measurements as well as models generated from historical building- and device-level data. We develop two implementations of the algorithm and conduct case studies using real demand data from households and commercial buildings to investigate the effectiveness of the algorithm. The case studies demonstrate that DFS can effectively perform online disaggregation and the choice and construction of models included in the algorithm affects its accuracy, which is comparable to that of a set of Kalman filters.
Tasks
Published	2017-01-16
URL	http://arxiv.org/abs/1701.04389v3
PDF	http://arxiv.org/pdf/1701.04389v3.pdf
PWC	https://paperswithcode.com/paper/real-time-energy-disaggregation-of-a
Repo
Framework

Recognizing Plans by Learning Embeddings from Observed Action Distributions


Title	Recognizing Plans by Learning Embeddings from Observed Action Distributions
Authors	Yantian Zha, Yikang Li, Sriram Gopalakrishnan, Baoxin Li, Subbarao Kambhampati
Abstract	Recent advances in visual activity recognition have raised the possibility of applications such as automated video surveillance. Effective approaches for such problems however require the ability to recognize the plans of agents from video information. Although traditional plan recognition algorithms depend on access to sophisticated planning domain models, one recent promising direction involves learning approximated (or shallow) domain models directly from the observed activity sequences DUP. One limitation is that such approaches expect observed action sequences as inputs. In many cases involving vision/sensing from raw data, there is considerable uncertainty about the specific action at any given time point. The most we can expect in such cases is probabilistic information about the action at that point. The input will then be sequences of such observed action distributions. In this work, we address the problem of constructing an effective data-interface that allows a plan recognition module to directly handle such observation distributions. Such an interface works like a bridge between the low-level perception module, and the high-level plan recognition module. We propose two approaches. The first involves resampling the distribution sequences to single action sequences, from which we could learn an action affinity model based on learned action (word) embeddings for plan recognition. The second is to directly learn action distribution embeddings by our proposed Distr2vec (distribution to vector) model, to construct an affinity model for plan recognition.
Tasks	Activity Recognition, Word Embeddings
Published	2017-12-05
URL	http://arxiv.org/abs/1712.01949v2
PDF	http://arxiv.org/pdf/1712.01949v2.pdf
PWC	https://paperswithcode.com/paper/recognizing-plans-by-learning-embeddings-from
Repo
Framework

Exploring Word Embeddings for Unsupervised Textual User-Generated Content Normalization


Title	Exploring Word Embeddings for Unsupervised Textual User-Generated Content Normalization
Authors	Thales Felipe Costa Bertaglia, Maria das Graças Volpe Nunes
Abstract	Text normalization techniques based on rules, lexicons or supervised training requiring large corpora are not scalable nor domain interchangeable, and this makes them unsuitable for normalizing user-generated content (UGC). Current tools available for Brazilian Portuguese make use of such techniques. In this work we propose a technique based on distributed representation of words (or word embeddings). It generates continuous numeric vectors of high-dimensionality to represent words. The vectors explicitly encode many linguistic regularities and patterns, as well as syntactic and semantic word relationships. Words that share semantic similarity are represented by similar vectors. Based on these features, we present a totally unsupervised, expandable and language and domain independent method for learning normalization lexicons from word embeddings. Our approach obtains high correction rate of orthographic errors and internet slang in product reviews, outperforming the current available tools for Brazilian Portuguese.
Tasks	Semantic Similarity, Semantic Textual Similarity, Word Embeddings
Published	2017-04-10
URL	http://arxiv.org/abs/1704.02963v1
PDF	http://arxiv.org/pdf/1704.02963v1.pdf
PWC	https://paperswithcode.com/paper/exploring-word-embeddings-for-unsupervised
Repo
Framework

Deep Radial Kernel Networks: Approximating Radially Symmetric Functions with Deep Networks


Title	Deep Radial Kernel Networks: Approximating Radially Symmetric Functions with Deep Networks
Authors	Brendan McCane, Lech Szymanski
Abstract	We prove that a particular deep network architecture is more efficient at approximating radially symmetric functions than the best known 2 or 3 layer networks. We use this architecture to approximate Gaussian kernel SVMs, and subsequently improve upon them with further training. The architecture and initial weights of the Deep Radial Kernel Network are completely specified by the SVM and therefore sidesteps the problem of empirically choosing an appropriate deep network architecture.
Tasks
Published	2017-03-09
URL	http://arxiv.org/abs/1703.03470v1
PDF	http://arxiv.org/pdf/1703.03470v1.pdf
PWC	https://paperswithcode.com/paper/deep-radial-kernel-networks-approximating
Repo
Framework

Thompson Sampling For Stochastic Bandits with Graph Feedback


Title	Thompson Sampling For Stochastic Bandits with Graph Feedback
Authors	Aristide C. Y. Tossou, Christos Dimitrakakis, Devdatt Dubhashi
Abstract	We present a novel extension of Thompson Sampling for stochastic sequential decision problems with graph feedback, even when the graph structure itself is unknown and/or changing. We provide theoretical guarantees on the Bayesian regret of the algorithm, linking its performance to the underlying properties of the graph. Thompson Sampling has the advantage of being applicable without the need to construct complicated upper confidence bounds for different problems. We illustrate its performance through extensive experimental results on real and simulated networks with graph feedback. More specifically, we tested our algorithms on power law, planted partitions and Erdo’s-Renyi graphs, as well as on graphs derived from Facebook and Flixster data. These all show that our algorithms clearly outperform related methods that employ upper confidence bounds, even if the latter use more information about the graph.
Tasks
Published	2017-01-16
URL	http://arxiv.org/abs/1701.04238v1
PDF	http://arxiv.org/pdf/1701.04238v1.pdf
PWC	https://paperswithcode.com/paper/thompson-sampling-for-stochastic-bandits-with
Repo
Framework

Exploration of Proximity Heuristics in Length Normalization


Title	Exploration of Proximity Heuristics in Length Normalization
Authors	Pranav Agrawal
Abstract	Ranking functions used in information retrieval are primarily used in the search engines and they are often adopted for various language processing applications. However, features used in the construction of ranking functions should be analyzed before applying it on a data set. This paper gives guidelines on construction of generalized ranking functions with application-dependent features. The paper prescribes a specific case of a generalized function for recommendation system using feature engineering guidelines on the given data set. The behavior of both generalized and specific functions are studied and implemented on the unstructured textual data. The proximity feature based ranking function has outperformed by 52% from regular BM25.
Tasks	Feature Engineering, Information Retrieval
Published	2017-01-05
URL	http://arxiv.org/abs/1701.01417v1
PDF	http://arxiv.org/pdf/1701.01417v1.pdf
PWC	https://paperswithcode.com/paper/exploration-of-proximity-heuristics-in-length
Repo
Framework

Generalizing Jensen and Bregman divergences with comparative convexity and the statistical Bhattacharyya distances with comparable means


Title	Generalizing Jensen and Bregman divergences with comparative convexity and the statistical Bhattacharyya distances with comparable means
Authors	Frank Nielsen, Richard Nock
Abstract	Comparative convexity is a generalization of convexity relying on abstract notions of means. We define the Jensen divergence and the Jensen diversity from the viewpoint of comparative convexity, and show how to obtain the generalized Bregman divergences as limit cases of skewed Jensen divergences. In particular, we report explicit formula of these generalized Bregman divergences when considering quasi-arithmetic means. Finally, we introduce a generalization of the Bhattacharyya statistical distances based on comparative means using relative convexity.
Tasks
Published	2017-02-16
URL	http://arxiv.org/abs/1702.04877v2
PDF	http://arxiv.org/pdf/1702.04877v2.pdf
PWC	https://paperswithcode.com/paper/generalizing-jensen-and-bregman-divergences
Repo
Framework

Understanding Early Word Learning in Situated Artificial Agents


Title	Understanding Early Word Learning in Situated Artificial Agents
Authors	Felix Hill, Stephen Clark, Karl Moritz Hermann, Phil Blunsom
Abstract	Neural network-based systems can now learn to locate the referents of words and phrases in images, answer questions about visual scenes, and execute symbolic instructions as first-person actors in partially-observable worlds. To achieve this so-called grounded language learning, models must overcome challenges that infants face when learning their first words. While it is notable that models with no meaningful prior knowledge overcome these obstacles, researchers currently lack a clear understanding of how they do so, a problem that we attempt to address in this paper. For maximum control and generality, we focus on a simple neural network-based language learning agent, trained via policy-gradient methods, which can interpret single-word instructions in a simulated 3D world. Whilst the goal is not to explicitly model infant word learning, we take inspiration from experimental paradigms in developmental psychology and apply some of these to the artificial agent, exploring the conditions under which established human biases and learning effects emerge. We further propose a novel method for visualising semantic representations in the agent.
Tasks	Policy Gradient Methods
Published	2017-10-26
URL	https://arxiv.org/abs/1710.09867v2
PDF	https://arxiv.org/pdf/1710.09867v2.pdf
PWC	https://paperswithcode.com/paper/understanding-grounded-language-learning
Repo
Framework