May 5, 2019

2889 words 14 mins read

Paper Group ANR 462

A Nonparametric Framework for Quantifying Generative Inference on Neuromorphic Systems. Multi-Task Curriculum Transfer Deep Learning of Clothing Attributes. Quantum Cognition Beyond Hilbert Space I: Fundamentals. Annotating Derivations: A New Evaluation Strategy and Dataset for Algebra Word Problems. A Truthful Mechanism with Biparameter Learning f …

A Nonparametric Framework for Quantifying Generative Inference on Neuromorphic Systems


Title	A Nonparametric Framework for Quantifying Generative Inference on Neuromorphic Systems
Authors	Ojash Neopane, Srinjoy Das, Ery Arias-Castro, Kenneth Kreutz-Delgado
Abstract	Restricted Boltzmann Machines and Deep Belief Networks have been successfully used in probabilistic generative model applications such as image occlusion removal, pattern completion and motion synthesis. Generative inference in such algorithms can be performed very efficiently on hardware using a Markov Chain Monte Carlo procedure called Gibbs sampling, where stochastic samples are drawn from noisy integrate and fire neurons implemented on neuromorphic substrates. Currently, no satisfactory metrics exist for evaluating the generative performance of such algorithms implemented on high-dimensional data for neuromorphic platforms. This paper demonstrates the application of nonparametric goodness-of-fit testing to both quantify the generative performance as well as provide decision-directed criteria for choosing the parameters of the neuromorphic Gibbs sampler and optimizing usage of hardware resources used during sampling.
Tasks
Published	2016-02-18
URL	http://arxiv.org/abs/1602.05996v1
PDF	http://arxiv.org/pdf/1602.05996v1.pdf
PWC	https://paperswithcode.com/paper/a-nonparametric-framework-for-quantifying
Repo
Framework

Multi-Task Curriculum Transfer Deep Learning of Clothing Attributes


Title	Multi-Task Curriculum Transfer Deep Learning of Clothing Attributes
Authors	Qi Dong, Shaogang Gong, Xiatian Zhu
Abstract	Recognising detailed clothing characteristics (fine-grained attributes) in unconstrained images of people in-the-wild is a challenging task for computer vision, especially when there is only limited training data from the wild whilst most data available for model learning are captured in well-controlled environments using fashion models (well lit, no background clutter, frontal view, high-resolution). In this work, we develop a deep learning framework capable of model transfer learning from well-controlled shop clothing images collected from web retailers to in-the-wild images from the street. Specifically, we formulate a novel Multi-Task Curriculum Transfer (MTCT) deep learning method to explore multiple sources of different types of web annotations with multi-labelled fine-grained attributes. Our multi-task loss function is designed to extract more discriminative representations in training by jointly learning all attributes, and our curriculum strategy exploits the staged easy-to-complex transfer learning motivated by cognitive studies. We demonstrate the advantages of the MTCT model over the state-of-the-art methods on the X-Domain benchmark, a large scale clothing attribute dataset. Moreover, we show that the MTCT model has a notable advantage over contemporary models when the training data size is small.
Tasks	Transfer Learning
Published	2016-10-12
URL	http://arxiv.org/abs/1610.03670v4
PDF	http://arxiv.org/pdf/1610.03670v4.pdf
PWC	https://paperswithcode.com/paper/multi-task-curriculum-transfer-deep-learning
Repo
Framework

Quantum Cognition Beyond Hilbert Space I: Fundamentals


Title	Quantum Cognition Beyond Hilbert Space I: Fundamentals
Authors	Diederik Aerts, Lyneth Beltran, Massimiliano Sassoli de Bianchi, Sandro Sozzo, Tomas Veloz
Abstract	The formalism of quantum theory in Hilbert space has been applied with success to the modeling and explanation of several cognitive phenomena, whereas traditional cognitive approaches were problematical. However, this ‘quantum cognition paradigm’ was recently challenged by its proven impossibility to simultaneously model ‘question order effects’ and ‘response replicability’. In Part I of this paper we describe sequential dichotomic measurements within an operational and realistic framework for human cognition elaborated by ourselves, and represent them in a quantum-like ‘extended Bloch representation’ where the Born rule of quantum probability does not necessarily hold. In Part II we apply this mathematical framework to successfully model question order effects, response replicability and unpacking effects, thus opening the way toward quantum cognition beyond Hilbert space.
Tasks
Published	2016-04-27
URL	http://arxiv.org/abs/1604.08268v1
PDF	http://arxiv.org/pdf/1604.08268v1.pdf
PWC	https://paperswithcode.com/paper/quantum-cognition-beyond-hilbert-space-i
Repo
Framework

Annotating Derivations: A New Evaluation Strategy and Dataset for Algebra Word Problems


Title	Annotating Derivations: A New Evaluation Strategy and Dataset for Algebra Word Problems
Authors	Shyam Upadhyay, Ming-Wei Chang
Abstract	We propose a new evaluation for automatic solvers for algebra word problems, which can identify mistakes that existing evaluations overlook. Our proposal is to evaluate such solvers using derivations, which reflect how an equation system was constructed from the word problem. To accomplish this, we develop an algorithm for checking the equivalence between two derivations, and show how derivation an- notations can be semi-automatically added to existing datasets. To make our experiments more comprehensive, we include the derivation annotation for DRAW-1K, a new dataset containing 1000 general algebra word problems. In our experiments, we found that the annotated derivations enable a more accurate evaluation of automatic solvers than previously used metrics. We release derivation annotations for over 2300 algebra word problems for future evaluations.
Tasks
Published	2016-09-23
URL	http://arxiv.org/abs/1609.07197v2
PDF	http://arxiv.org/pdf/1609.07197v2.pdf
PWC	https://paperswithcode.com/paper/annotating-derivations-a-new-evaluation
Repo
Framework

A Truthful Mechanism with Biparameter Learning for Online Crowdsourcing


Title	A Truthful Mechanism with Biparameter Learning for Online Crowdsourcing
Authors	Satyanath Bhat, Divya Padmanabhan, Shweta Jain, Y Narahari
Abstract	We study a problem of allocating divisible jobs, arriving online, to workers in a crowdsourcing setting which involves learning two parameters of strategically behaving workers. Each job is split into a certain number of tasks that are then allocated to workers. Each arriving job has to be completed within a deadline and each task has to be completed satisfying an upper bound on probability of failure. The job population is homogeneous while the workers are heterogeneous in terms of costs, completion times, and times to failure. The job completion time and time to failure of each worker are stochastic with fixed but unknown means. The requester is faced with the challenge of learning two separate parameters of each (strategically behaving) worker simultaneously, namely, the mean job completion time and the mean time to failure. The time to failure of a worker depends on the duration of the task handled by the worker. Assuming non-strategic workers to start with, we solve this biparameter learning problem by applying the Robust UCB algorithm. Then, we non-trivially extend this algorithm to the setting where the workers are strategic about their costs. Our proposed mechanism is dominant strategy incentive compatible and ex-post individually rational with asymptotically optimal regret performance.
Tasks
Published	2016-02-12
URL	http://arxiv.org/abs/1602.04032v1
PDF	http://arxiv.org/pdf/1602.04032v1.pdf
PWC	https://paperswithcode.com/paper/a-truthful-mechanism-with-biparameter
Repo
Framework

Toward Mention Detection Robustness with Recurrent Neural Networks


Title	Toward Mention Detection Robustness with Recurrent Neural Networks
Authors	Thien Huu Nguyen, Avirup Sil, Georgiana Dinu, Radu Florian
Abstract	One of the key challenges in natural language processing (NLP) is to yield good performance across application domains and languages. In this work, we investigate the robustness of the mention detection systems, one of the fundamental tasks in information extraction, via recurrent neural networks (RNNs). The advantage of RNNs over the traditional approaches is their capacity to capture long ranges of context and implicitly adapt the word embeddings, trained on a large corpus, into a task-specific word representation, but still preserve the original semantic generalization to be helpful across domains. Our systematic evaluation for RNN architectures demonstrates that RNNs not only outperform the best reported systems (up to 9% relative error reduction) in the general setting but also achieve the state-of-the-art performance in the cross-domain setting for English. Regarding other languages, RNNs are significantly better than the traditional methods on the similar task of named entity recognition for Dutch (up to 22% relative error reduction).
Tasks	Named Entity Recognition, Word Embeddings
Published	2016-02-24
URL	http://arxiv.org/abs/1602.07749v1
PDF	http://arxiv.org/pdf/1602.07749v1.pdf
PWC	https://paperswithcode.com/paper/toward-mention-detection-robustness-with
Repo
Framework

Rank Correlation Measure: A Representational Transformation for Biometric Template Protection


Title	Rank Correlation Measure: A Representational Transformation for Biometric Template Protection
Authors	Zhe Jin, Yen-Lung Lai, Andrew Beng Jin Teoh
Abstract	Despite a variety of theoretical-sound techniques have been proposed for biometric template protection, there is rarely practical solution that guarantees non-invertibility, cancellability, non-linkability and performance simultaneously. In this paper, a ranking-based representational transformation is proposed for fingerprint templates. The proposed method transforms a real-valued feature vector into index code such that the pairwise-order measure in the resultant codes are closely correlated with rank similarity measure. Such a ranking based technique offers two major merits: 1) Resilient to noises/perturbations in numeric values; and 2) Highly nonlinear embedding based on partial order statistics. The former takes care of the accuracy performance mitigating numeric noises/perturbations while the latter offers strong non-invertible transformation via nonlinear feature embedding from Euclidean to Rank space that leads to toughness in inversion. The experimental results demonstrate reasonable accuracy performance on benchmark FVC2002 and FVC2004 fingerprint databases, thus confirm the proposition of the rank correlation. Moreover, the security and privacy analysis justify the strong capability against the existing major privacy attacks.
Tasks
Published	2016-07-23
URL	http://arxiv.org/abs/1607.06902v1
PDF	http://arxiv.org/pdf/1607.06902v1.pdf
PWC	https://paperswithcode.com/paper/rank-correlation-measure-a-representational
Repo
Framework

Cluster-Wise Ratio Tests for Fast Camera Localization


Title	Cluster-Wise Ratio Tests for Fast Camera Localization
Authors	Raúl Díaz, Charless C. Fowlkes
Abstract	Feature point matching for camera localization suffers from scalability problems. Even when feature descriptors associated with 3D scene points are locally unique, as coverage grows, similar or repeated features become increasingly common. As a result, the standard distance ratio-test used to identify reliable image feature points is overly restrictive and rejects many good candidate matches. We propose a simple coarse-to-fine strategy that uses conservative approximations to robust local ratio-tests that can be computed efficiently using global approximate k-nearest neighbor search. We treat these forward matches as votes in camera pose space and use them to prioritize back-matching within candidate camera pose clusters, exploiting feature co-visibility captured by clustering the 3D model camera pose graph. This approach achieves state-of-the-art camera localization results on a variety of popular benchmarks, outperforming several methods that use more complicated data structures and that make more restrictive assumptions on camera pose. We also carry out diagnostic analyses on a difficult test dataset containing globally repetitive structure that suggest our approach successfully adapts to the challenges of large-scale image localization.
Tasks	Camera Localization
Published	2016-12-06
URL	http://arxiv.org/abs/1612.01689v2
PDF	http://arxiv.org/pdf/1612.01689v2.pdf
PWC	https://paperswithcode.com/paper/cluster-wise-ratio-tests-for-fast-camera
Repo
Framework

CNN-LTE: a Class of 1-X Pooling Convolutional Neural Networks on Label Tree Embeddings for Audio Scene Recognition


Title	CNN-LTE: a Class of 1-X Pooling Convolutional Neural Networks on Label Tree Embeddings for Audio Scene Recognition
Authors	Huy Phan, Lars Hertel, Marco Maass, Philipp Koch, Alfred Mertins
Abstract	We describe in this report our audio scene recognition system submitted to the DCASE 2016 challenge. Firstly, given the label set of the scenes, a label tree is automatically constructed. This category taxonomy is then used in the feature extraction step in which an audio scene instance is represented by a label tree embedding image. Different convolutional neural networks, which are tailored for the task at hand, are finally learned on top of the image features for scene recognition. Our system reaches an overall recognition accuracy of 81.2% and 83.3% and outperforms the DCASE 2016 baseline with absolute improvements of 8.7% and 6.1% on the development and test data, respectively.
Tasks	Scene Recognition
Published	2016-07-08
URL	http://arxiv.org/abs/1607.02303v2
PDF	http://arxiv.org/pdf/1607.02303v2.pdf
PWC	https://paperswithcode.com/paper/cnn-lte-a-class-of-1-x-pooling-convolutional
Repo
Framework

Utilização de Grafos e Matriz de Similaridade na Sumarização Automática de Documentos Baseada em Extração de Frases


Title	Utilização de Grafos e Matriz de Similaridade na Sumarização Automática de Documentos Baseada em Extração de Frases
Authors	Elvys Linhares Pontes
Abstract	The internet increased the amount of information available. However, the reading and understanding of this information are costly tasks. In this scenario, the Natural Language Processing (NLP) applications enable very important solutions, highlighting the Automatic Text Summarization (ATS), which produce a summary from one or more source texts. Automatically summarizing one or more texts, however, is a complex task because of the difficulties inherent to the analysis and generation of this summary. This master’s thesis describes the main techniques and methodologies (NLP and heuristics) to generate summaries. We have also addressed and proposed some heuristics based on graphs and similarity matrix to measure the relevance of judgments and to generate summaries by extracting sentences. We used the multiple languages (English, French and Spanish), CSTNews (Brazilian Portuguese), RPM (French) and DECODA (French) corpus to evaluate the developped systems. The results obtained were quite interesting.
Tasks	Text Summarization
Published	2016-02-05
URL	http://arxiv.org/abs/1602.02047v1
PDF	http://arxiv.org/pdf/1602.02047v1.pdf
PWC	https://paperswithcode.com/paper/utilizacao-de-grafos-e-matriz-de-similaridade
Repo
Framework

A backward pass through a CNN using a generative model of its activations


Title	A backward pass through a CNN using a generative model of its activations
Authors	Huayan Wang, Anna Chen, Yi Liu, Dileep George, D. Scott Phoenix
Abstract	Neural networks have shown to be a practical way of building a very complex mapping between a pre-specified input space and output space. For example, a convolutional neural network (CNN) mapping an image into one of a thousand object labels is approaching human performance in this particular task. However the mapping (neural network) does not automatically lend itself to other forms of queries, for example, to detect/reconstruct object instances, to enforce top-down signal on ambiguous inputs, or to recover object instances from occlusion. One way to address these queries is a backward pass through the network that fuses top-down and bottom-up information. In this paper, we show a way of building such a backward pass by defining a generative model of the neural network’s activations. Approximate inference of the model would naturally take the form of a backward pass through the CNN layers, and it addresses the aforementioned queries in a unified framework.
Tasks
Published	2016-11-08
URL	http://arxiv.org/abs/1611.02767v1
PDF	http://arxiv.org/pdf/1611.02767v1.pdf
PWC	https://paperswithcode.com/paper/a-backward-pass-through-a-cnn-using-a
Repo
Framework

A Supervised Authorship Attribution Framework for Bengali Language


Title	A Supervised Authorship Attribution Framework for Bengali Language
Authors	Shanta Phani, Shibamouli Lahiri, Arindam Biswas
Abstract	Authorship Attribution is a long-standing problem in Natural Language Processing. Several statistical and computational methods have been used to find a solution to this problem. In this paper, we have proposed methods to deal with the authorship attribution problem in Bengali.
Tasks
Published	2016-07-13
URL	http://arxiv.org/abs/1607.05650v2
PDF	http://arxiv.org/pdf/1607.05650v2.pdf
PWC	https://paperswithcode.com/paper/a-supervised-authorship-attribution-framework
Repo
Framework

Unsupervised Feature Selection Based on the Morisita Estimator of Intrinsic Dimension


Title	Unsupervised Feature Selection Based on the Morisita Estimator of Intrinsic Dimension
Authors	Jean Golay, Mikhail Kanevski
Abstract	This paper deals with a new filter algorithm for selecting the smallest subset of features carrying all the information content of a data set (i.e. for removing redundant features). It is an advanced version of the fractal dimension reduction technique, and it relies on the recently introduced Morisita estimator of Intrinsic Dimension (ID). Here, the ID is used to quantify dependencies between subsets of features, which allows the effective processing of highly non-linear data. The proposed algorithm is successfully tested on simulated and real world case studies. Different levels of sample size and noise are examined along with the variability of the results. In addition, a comprehensive procedure based on random forests shows that the data dimensionality is significantly reduced by the algorithm without loss of relevant information. And finally, comparisons with benchmark feature selection techniques demonstrate the promising performance of this new filter.
Tasks	Dimensionality Reduction, Feature Selection
Published	2016-08-19
URL	http://arxiv.org/abs/1608.05581v5
PDF	http://arxiv.org/pdf/1608.05581v5.pdf
PWC	https://paperswithcode.com/paper/unsupervised-feature-selection-based-on-the
Repo
Framework

Self-expressive Dictionary Learning for Dynamic 3D Reconstruction


Title	Self-expressive Dictionary Learning for Dynamic 3D Reconstruction
Authors	Enliang Zheng, Dinghuang Ji, Enrique Dunn, Jan-Michael Frahm
Abstract	We target the problem of sparse 3D reconstruction of dynamic objects observed by multiple unsynchronized video cameras with unknown temporal overlap. To this end, we develop a framework to recover the unknown structure without sequencing information across video sequences. Our proposed compressed sensing framework poses the estimation of 3D structure as the problem of dictionary learning, where the dictionary is defined as an aggregation of the temporally varying 3D structures. Given the smooth motion of dynamic objects, we observe any element in the dictionary can be well approximated by a sparse linear combination of other elements in the same dictionary (i. e. self-expression). Moreover, the sparse coefficients describing a locally linear 3D structural interpolation reveal the local sequencing information. Our formulation optimizes a biconvex cost function that leverages a compressed sensing formulation and enforces both structural dependency coherence across video streams, as well as motion smoothness across estimates from common video sources. We further analyze the reconstructability of our approach under different capture scenarios, and its comparison and relation to existing methods. Experimental results on large amounts of synthetic data as well as real imagery demonstrate the effectiveness of our approach.
Tasks	3D Reconstruction, Dictionary Learning
Published	2016-05-22
URL	http://arxiv.org/abs/1605.06863v1
PDF	http://arxiv.org/pdf/1605.06863v1.pdf
PWC	https://paperswithcode.com/paper/self-expressive-dictionary-learning-for
Repo
Framework

Variational inference for Monte Carlo objectives


Title	Variational inference for Monte Carlo objectives
Authors	Andriy Mnih, Danilo J. Rezende
Abstract	Recent progress in deep latent variable models has largely been driven by the development of flexible and scalable variational inference methods. Variational training of this type involves maximizing a lower bound on the log-likelihood, using samples from the variational posterior to compute the required gradients. Recently, Burda et al. (2016) have derived a tighter lower bound using a multi-sample importance sampling estimate of the likelihood and showed that optimizing it yields models that use more of their capacity and achieve higher likelihoods. This development showed the importance of such multi-sample objectives and explained the success of several related approaches. We extend the multi-sample approach to discrete latent variables and analyze the difficulty encountered when estimating the gradients involved. We then develop the first unbiased gradient estimator designed for importance-sampled objectives and evaluate it at training generative and structured output prediction models. The resulting estimator, which is based on low-variance per-sample learning signals, is both simpler and more effective than the NVIL estimator proposed for the single-sample variational objective, and is competitive with the currently used biased estimators.
Tasks	Latent Variable Models
Published	2016-02-22
URL	http://arxiv.org/abs/1602.06725v2
PDF	http://arxiv.org/pdf/1602.06725v2.pdf
PWC	https://paperswithcode.com/paper/variational-inference-for-monte-carlo
Repo
Framework