May 6, 2019

3185 words 15 mins read

Paper Group ANR 410

Automated Alertness and Emotion Detection for Empathic Feedback During E-Learning. Eye detection in digital images: challenges and solutions. Improving Neural Network Generalization by Combining Parallel Circuits with Dropout. DISCO Nets: DISsimilarity COefficient Networks. Linear Convergence of SVRG in Statistical Estimation. Whiteout: Gaussian Ad …

Automated Alertness and Emotion Detection for Empathic Feedback During E-Learning


Title	Automated Alertness and Emotion Detection for Empathic Feedback During E-Learning
Authors	S L Happy, A. Dasgupta, P. Patnaik, A. Routray
Abstract	In the context of education technology, empathic interaction with the user and feedback by the learning system using multiple inputs such as video, voice and text inputs is an important area of research. In this paper, a nonintrusive, standalone model for intelligent assessment of alertness and emotional state as well as generation of appropriate feedback has been proposed. Using the non-intrusive visual cues, the system classifies emotion and alertness state of the user, and provides appropriate feedback according to the detected cognitive state using facial expressions, ocular parameters, postures, and gestures. Assessment of alertness level using ocular parameters such as PERCLOS and saccadic parameters, emotional state from facial expression analysis, and detection of both relevant cognitive and emotional states from upper body gestures and postures has been proposed. Integration of such a system in e-learning environment is expected to enhance students performance through interaction, feedback, and positive mood induction.
Tasks
Published	2016-04-01
URL	http://arxiv.org/abs/1604.00312v1
PDF	http://arxiv.org/pdf/1604.00312v1.pdf
PWC	https://paperswithcode.com/paper/automated-alertness-and-emotion-detection-for
Repo
Framework

Eye detection in digital images: challenges and solutions


Title	Eye detection in digital images: challenges and solutions
Authors	Mitra Montazeri, Mahdieh Montazeri, Saeid Saryazdi
Abstract	Eye Detection has an important role in the field of biometric identification and known as one method of person’s identification. In recent years, many efforts have been done which can detect eye automatically and with different image conditions. However, each method has its own drawbacks which can control some of these conditions. In this paper, different methods of eye detection will be categorized and explained. In each category, the advantages and disadvantages of each method will be presented.
Tasks
Published	2016-01-19
URL	http://arxiv.org/abs/1601.04871v2
PDF	http://arxiv.org/pdf/1601.04871v2.pdf
PWC	https://paperswithcode.com/paper/eye-detection-in-digital-images-challenges
Repo
Framework

Improving Neural Network Generalization by Combining Parallel Circuits with Dropout


Title	Improving Neural Network Generalization by Combining Parallel Circuits with Dropout
Authors	Kien Tuong Phan, Tomas Henrique Maul, Tuong Thuy Vu, Lai Weng Kin
Abstract	In an attempt to solve the lengthy training times of neural networks, we proposed Parallel Circuits (PCs), a biologically inspired architecture. Previous work has shown that this approach fails to maintain generalization performance in spite of achieving sharp speed gains. To address this issue, and motivated by the way Dropout prevents node co-adaption, in this paper, we suggest an improvement by extending Dropout to the PC architecture. The paper provides multiple insights into this combination, including a variety of fusion approaches. Experiments show promising results in which improved error rates are achieved in most cases, whilst maintaining the speed advantage of the PC approach.
Tasks
Published	2016-12-15
URL	http://arxiv.org/abs/1612.04970v1
PDF	http://arxiv.org/pdf/1612.04970v1.pdf
PWC	https://paperswithcode.com/paper/improving-neural-network-generalization-by
Repo
Framework

DISCO Nets: DISsimilarity COefficient Networks


Title	DISCO Nets: DISsimilarity COefficient Networks
Authors	Diane Bouchacourt, M. Pawan Kumar, Sebastian Nowozin
Abstract	We present a new type of probabilistic model which we call DISsimilarity COefficient Networks (DISCO Nets). DISCO Nets allow us to efficiently sample from a posterior distribution parametrised by a neural network. During training, DISCO Nets are learned by minimising the dissimilarity coefficient between the true distribution and the estimated distribution. This allows us to tailor the training to the loss related to the task at hand. We empirically show that (i) by modeling uncertainty on the output value, DISCO Nets outperform equivalent non-probabilistic predictive networks and (ii) DISCO Nets accurately model the uncertainty of the output, outperforming existing probabilistic models based on deep neural networks.
Tasks
Published	2016-06-08
URL	http://arxiv.org/abs/1606.02556v5
PDF	http://arxiv.org/pdf/1606.02556v5.pdf
PWC	https://paperswithcode.com/paper/disco-nets-dissimilarity-coefficient-networks
Repo
Framework

Linear Convergence of SVRG in Statistical Estimation


Title	Linear Convergence of SVRG in Statistical Estimation
Authors	Chao Qu, Yan Li, Huan Xu
Abstract	SVRG and its variants are among the state of art optimization algorithms for large scale machine learning problems. It is well known that SVRG converges linearly when the objective function is strongly convex. However this setup can be restrictive, and does not include several important formulations such as Lasso, group Lasso, logistic regression, and some non-convex models including corrected Lasso and SCAD. In this paper, we prove that, for a class of statistical M-estimators covering examples mentioned above, SVRG solves the formulation with {\em a linear convergence rate} without strong convexity or even convexity. Our analysis makes use of {\em restricted strong convexity}, under which we show that SVRG converges linearly to the fundamental statistical precision of the model, i.e., the difference between true unknown parameter $\theta^*$ and the optimal solution $\hat{\theta}$ of the model.
Tasks
Published	2016-11-07
URL	http://arxiv.org/abs/1611.01957v3
PDF	http://arxiv.org/pdf/1611.01957v3.pdf
PWC	https://paperswithcode.com/paper/linear-convergence-of-svrg-in-statistical
Repo
Framework

Whiteout: Gaussian Adaptive Noise Regularization in Deep Neural Networks


Title	Whiteout: Gaussian Adaptive Noise Regularization in Deep Neural Networks
Authors	Yinan Li, Fang Liu
Abstract	Noise injection (NI) is an efficient technique to mitigate over-fitting in neural networks (NNs). The Bernoulli NI procedure as implemented in dropout and shakeout has connections with $l_1$ and $l_2$ regularization for the NN model parameters. We propose whiteout, a family NI regularization techniques (NIRT) through injecting adaptive Gaussian noises during the training of NNs. Whiteout is the first NIRT than imposes a broad range of the $l_{\gamma}$ sparsity regularization $(\gamma\in(0,2))$ without having to involving the $l_2$ regularization. Whiteout can also be extended to offer regularizations similar to the adaptive lasso and group lasso. We establish the regularization effect of whiteout in the framework of generalized linear models with closed-form penalty terms and show that whiteout stabilizes the training of NNs with decreased sensitivity to small perturbations in the input. We establish that the noise-perturbed empirical loss function (pelf) with whiteout converges almost surely to the ideal loss function (ilf), and the minimizer of the pelf is consistent for the minimizer of the ilf. We derive the tail bound on the pelf to establish the practical feasibility in its minimization. The superiority of whiteout over the Bernoulli NIRTs, dropout and shakeout, in learning NNs with relatively small-sized training sets and non-inferiority in large-sized training sets is demonstrated in both simulated and real-life data sets. This work represents the first in-depth theoretical, methodological, and practical examination of the regularization effects of both additive and multiplicative Gaussian NI in deep NNs.
Tasks
Published	2016-12-05
URL	http://arxiv.org/abs/1612.01490v4
PDF	http://arxiv.org/pdf/1612.01490v4.pdf
PWC	https://paperswithcode.com/paper/whiteout-gaussian-adaptive-noise
Repo
Framework

The iterative reweighted Mixed-Norm Estimate for spatio-temporal MEG/EEG source reconstruction


Title	The iterative reweighted Mixed-Norm Estimate for spatio-temporal MEG/EEG source reconstruction
Authors	Daniel Strohmeier, Yousra Bekhti, Jens Haueisen, Alexandre Gramfort
Abstract	Source imaging based on magnetoencephalography (MEG) and electroencephalography (EEG) allows for the non-invasive analysis of brain activity with high temporal and good spatial resolution. As the bioelectromagnetic inverse problem is ill-posed, constraints are required. For the analysis of evoked brain activity, spatial sparsity of the neuronal activation is a common assumption. It is often taken into account using convex constraints based on the l1-norm. The resulting source estimates are however biased in amplitude and often suboptimal in terms of source selection due to high correlations in the forward model. In this work, we demonstrate that an inverse solver based on a block-separable penalty with a Frobenius norm per block and a l0.5-quasinorm over blocks addresses both of these issues. For solving the resulting non-convex optimization problem, we propose the iterative reweighted Mixed Norm Estimate (irMxNE), an optimization scheme based on iterative reweighted convex surrogate optimization problems, which are solved efficiently using a block coordinate descent scheme and an active set strategy. We compare the proposed sparse imaging method to the dSPM and the RAP-MUSIC approach based on two MEG data sets. We provide empirical evidence based on simulations and analysis of MEG data that the proposed method improves on the standard Mixed Norm Estimate (MxNE) in terms of amplitude bias, support recovery, and stability.
Tasks	EEG
Published	2016-07-28
URL	http://arxiv.org/abs/1607.08458v1
PDF	http://arxiv.org/pdf/1607.08458v1.pdf
PWC	https://paperswithcode.com/paper/the-iterative-reweighted-mixed-norm-estimate
Repo
Framework

Core Sampling Framework for Pixel Classification


Title	Core Sampling Framework for Pixel Classification
Authors	Manohar Karki, Robert DiBiano, Saikat Basu, Supratik Mukhopadhyay
Abstract	The intermediate map responses of a Convolutional Neural Network (CNN) contain information about an image that can be used to extract contextual knowledge about it. In this paper, we present a core sampling framework that is able to use these activation maps from several layers as features to another neural network using transfer learning to provide an understanding of an input image. Our framework creates a representation that combines features from the test data and the contextual knowledge gained from the responses of a pretrained network, processes it and feeds it to a separate Deep Belief Network. We use this representation to extract more information from an image at the pixel level, hence gaining understanding of the whole image. We experimentally demonstrate the usefulness of our framework using a pretrained VGG-16 model to perform segmentation on the BAERI dataset of Synthetic Aperture Radar(SAR) imagery and the CAMVID dataset.
Tasks	Transfer Learning
Published	2016-12-06
URL	http://arxiv.org/abs/1612.01981v1
PDF	http://arxiv.org/pdf/1612.01981v1.pdf
PWC	https://paperswithcode.com/paper/core-sampling-framework-for-pixel
Repo
Framework

Multi-Label Zero-Shot Learning via Concept Embedding


Title	Multi-Label Zero-Shot Learning via Concept Embedding
Authors	Ubai Sandouk, Ke Chen
Abstract	Zero Shot Learning (ZSL) enables a learning model to classify instances of an unseen class during training. While most research in ZSL focuses on single-label classification, few studies have been done in multi-label ZSL, where an instance is associated with a set of labels simultaneously, due to the difficulty in modeling complex semantics conveyed by a set of labels. In this paper, we propose a novel approach to multi-label ZSL via concept embedding learned from collections of public users’ annotations of multimedia. Thanks to concept embedding, multi-label ZSL can be done by efficiently mapping an instance input features onto the concept embedding space in a similar manner used in single-label ZSL. Moreover, our semantic learning model is capable of embedding an out-of-vocabulary label by inferring its meaning from its co-occurring labels. Thus, our approach allows both seen and unseen labels during the concept embedding learning to be used in the aforementioned instance mapping, which makes multi-label ZSL more flexible and suitable for real applications. Experimental results of multi-label ZSL on images and music tracks suggest that our approach outperforms a state-of-the-art multi-label ZSL model and can deal with a scenario involving out-of-vocabulary labels without re-training the semantics learning model.
Tasks	Zero-Shot Learning
Published	2016-06-01
URL	http://arxiv.org/abs/1606.00282v1
PDF	http://arxiv.org/pdf/1606.00282v1.pdf
PWC	https://paperswithcode.com/paper/multi-label-zero-shot-learning-via-concept
Repo
Framework

Regularization Parameter Selection for a Bayesian Multi-Level Group Lasso Regression Model with Application to Imaging Genomics


Title	Regularization Parameter Selection for a Bayesian Multi-Level Group Lasso Regression Model with Application to Imaging Genomics
Authors	Farouk S. Nathoo, Keelin Greenlaw, Mary Lesperance
Abstract	We investigate the choice of tuning parameters for a Bayesian multi-level group lasso model developed for the joint analysis of neuroimaging and genetic data. The regression model we consider relates multivariate phenotypes consisting of brain summary measures (volumetric and cortical thickness values) to single nucleotide polymorphism (SNPs) data and imposes penalization at two nested levels, the first corresponding to genes and the second corresponding to SNPs. Associated with each level in the penalty is a tuning parameter which corresponds to a hyperparameter in the hierarchical Bayesian formulation. Following previous work on Bayesian lassos we consider the estimation of tuning parameters through either hierarchical Bayes based on hyperpriors and Gibbs sampling or through empirical Bayes based on maximizing the marginal likelihood using a Monte Carlo EM algorithm. For the specific model under consideration we find that these approaches can lead to severe overshrinkage of the regression parameter estimates in the high-dimensional setting or when the genetic effects are weak. We demonstrate these problems through simulation examples and study an approximation to the marginal likelihood which sheds light on the cause of this problem. We then suggest an alternative approach based on the widely applicable information criterion (WAIC), an asymptotic approximation to leave-one-out cross-validation that can be computed conveniently within an MCMC framework.
Tasks
Published	2016-03-27
URL	http://arxiv.org/abs/1603.08163v1
PDF	http://arxiv.org/pdf/1603.08163v1.pdf
PWC	https://paperswithcode.com/paper/regularization-parameter-selection-for-a
Repo
Framework

Hybrid Approach for Single Text Document Summarization using Statistical and Sentiment Features


Title	Hybrid Approach for Single Text Document Summarization using Statistical and Sentiment Features
Authors	Chandra Shekhar Yadav, Aditi Sharan
Abstract	Summarization is a way to represent same information in concise way with equal sense. This can be categorized in two type Abstractive and Extractive type. Our work is focused around Extractive summarization. A generic approach to extractive summarization is to consider sentence as an entity, score each sentence based on some indicative features to ascertain the quality of sentence for inclusion in summary. Sort the sentences on the score and consider top n sentences for summarization. Mostly statistical features have been used for scoring the sentences. We are proposing a hybrid model for a single text document summarization. This hybrid model is an extraction based approach, which is combination of Statistical and semantic technique. The hybrid model depends on the linear combination of statistical measures : sentence position, TF-IDF, Aggregate similarity, centroid, and semantic measure. Our idea to include sentiment analysis for salient sentence extraction is derived from the concept that emotion plays an important role in communication to effectively convey any message hence, it can play a vital role in text document summarization. For comparison we have generated five system summaries Proposed Work, MEAD system, Microsoft system, OPINOSIS system, and Human generated summary, and evaluation is done using ROUGE score.
Tasks	Document Summarization, Sentiment Analysis
Published	2016-01-03
URL	https://arxiv.org/abs/1601.00643v1
PDF	https://arxiv.org/pdf/1601.00643v1.pdf
PWC	https://paperswithcode.com/paper/hybrid-approach-for-single-text-document
Repo
Framework

Proceedings of the 2016 Workshop on Semantic Spaces at the Intersection of NLP, Physics and Cognitive Science


Title	Proceedings of the 2016 Workshop on Semantic Spaces at the Intersection of NLP, Physics and Cognitive Science
Authors	Dimitrios Kartsaklis, Martha Lewis, Laura Rimell
Abstract	This volume contains the Proceedings of the 2016 Workshop on Semantic Spaces at the Intersection of NLP, Physics and Cognitive Science (SLPCS 2016), which was held on the 11th of June at the University of Strathclyde, Glasgow, and was co-located with Quantum Physics and Logic (QPL 2016). Exploiting the common ground provided by the concept of a vector space, the workshop brought together researchers working at the intersection of Natural Language Processing (NLP), cognitive science, and physics, offering them an appropriate forum for presenting their uniquely motivated work and ideas. The interplay between these three disciplines inspired theoretically motivated approaches to the understanding of how word meanings interact with each other in sentences and discourse, how diagrammatic reasoning depicts and simplifies this interaction, how language models are determined by input from the world, and how word and sentence meanings interact logically. This first edition of the workshop consisted of three invited talks from distinguished speakers (Hans Briegel, Peter G"ardenfors, Dominic Widdows) and eight presentations of selected contributed papers. Each submission was refereed by at least three members of the Programme Committee, who delivered detailed and insightful comments and suggestions.
Tasks
Published	2016-08-02
URL	http://arxiv.org/abs/1608.01018v1
PDF	http://arxiv.org/pdf/1608.01018v1.pdf
PWC	https://paperswithcode.com/paper/proceedings-of-the-2016-workshop-on-semantic
Repo
Framework

First-Person Activity Forecasting with Online Inverse Reinforcement Learning


Title	First-Person Activity Forecasting with Online Inverse Reinforcement Learning
Authors	Nicholas Rhinehart, Kris M. Kitani
Abstract	We address the problem of incrementally modeling and forecasting long-term goals of a first-person camera wearer: what the user will do, where they will go, and what goal they seek. In contrast to prior work in trajectory forecasting, our algorithm, DARKO, goes further to reason about semantic states (will I pick up an object?), and future goal states that are far in terms of both space and time. DARKO learns and forecasts from first-person visual observations of the user’s daily behaviors via an Online Inverse Reinforcement Learning (IRL) approach. Classical IRL discovers only the rewards in a batch setting, whereas DARKO discovers the states, transitions, rewards, and goals of a user from streaming data. Among other results, we show DARKO forecasts goals better than competing methods in both noisy and ideal settings, and our approach is theoretically and empirically no-regret.
Tasks
Published	2016-12-22
URL	http://arxiv.org/abs/1612.07796v3
PDF	http://arxiv.org/pdf/1612.07796v3.pdf
PWC	https://paperswithcode.com/paper/first-person-activity-forecasting-with-online
Repo
Framework

Median-Based Generation of Synthetic Speech Durations using a Non-Parametric Approach


Title	Median-Based Generation of Synthetic Speech Durations using a Non-Parametric Approach
Authors	Srikanth Ronanki, Oliver Watts, Simon King, Gustav Eje Henter
Abstract	This paper proposes a new approach to duration modelling for statistical parametric speech synthesis in which a recurrent statistical model is trained to output a phone transition probability at each timestep (acoustic frame). Unlike conventional approaches to duration modelling – which assume that duration distributions have a particular form (e.g., a Gaussian) and use the mean of that distribution for synthesis – our approach can in principle model any distribution supported on the non-negative integers. Generation from this model can be performed in many ways; here we consider output generation based on the median predicted duration. The median is more typical (more probable) than the conventional mean duration, is robust to training-data irregularities, and enables incremental generation. Furthermore, a frame-level approach to duration prediction is consistent with a longer-term goal of modelling durations and acoustic features together. Results indicate that the proposed method is competitive with baseline approaches in approximating the median duration of held-out natural speech.
Tasks	Speech Synthesis
Published	2016-08-22
URL	http://arxiv.org/abs/1608.06134v2
PDF	http://arxiv.org/pdf/1608.06134v2.pdf
PWC	https://paperswithcode.com/paper/median-based-generation-of-synthetic-speech
Repo
Framework

Feedforward Initialization for Fast Inference of Deep Generative Networks is biologically plausible


Title	Feedforward Initialization for Fast Inference of Deep Generative Networks is biologically plausible
Authors	Yoshua Bengio, Benjamin Scellier, Olexa Bilaniuk, Joao Sacramento, Walter Senn
Abstract	We consider deep multi-layered generative models such as Boltzmann machines or Hopfield nets in which computation (which implements inference) is both recurrent and stochastic, but where the recurrence is not to model sequential structure, only to perform computation. We find conditions under which a simple feedforward computation is a very good initialization for inference, after the input units are clamped to observed values. It means that after the feedforward initialization, the recurrent network is very close to a fixed point of the network dynamics, where the energy gradient is 0. The main condition is that consecutive layers form a good auto-encoder, or more generally that different groups of inputs into the unit (in particular, bottom-up inputs on one hand, top-down inputs on the other hand) are consistent with each other, producing the same contribution into the total weighted sum of inputs. In biological terms, this would correspond to having each dendritic branch correctly predicting the aggregate input from all the dendritic branches, i.e., the soma potential. This is consistent with the prediction that the synaptic weights into dendritic branches such as those of the apical and basal dendrites of pyramidal cells are trained to minimize the prediction error made by the dendritic branch when the target is the somatic activity. Whereas previous work has shown how to achieve fast negative phase inference (when the model is unclamped) in a predictive recurrent model, this contribution helps to achieve fast positive phase inference (when the target output is clamped) in such recurrent neural models.
Tasks
Published	2016-06-06
URL	http://arxiv.org/abs/1606.01651v2
PDF	http://arxiv.org/pdf/1606.01651v2.pdf
PWC	https://paperswithcode.com/paper/feedforward-initialization-for-fast-inference
Repo
Framework