May 6, 2019

2775 words 14 mins read

Paper Group ANR 351

Detecting Violence in Video using Subclasses. An Approach to Stable Gradient Descent Adaptation of Higher-Order Neural Units. Wikiometrics: A Wikipedia Based Ranking System. Asking the metaquestions in constraint tractability. Scaled stochastic gradient descent for low-rank matrix completion. Permutation NMF. M2CAI Workflow Challenge: Convolutional …

Detecting Violence in Video using Subclasses


Title	Detecting Violence in Video using Subclasses
Authors	Xirong Li, Yujia Huo, Jieping Xu, Qin Jin
Abstract	This paper attacks the challenging problem of violence detection in videos. Different from existing works focusing on combining multi-modal features, we go one step further by adding and exploiting subclasses visually related to violence. We enrich the MediaEval 2015 violence dataset by \emph{manually} labeling violence videos with respect to the subclasses. Such fine-grained annotations not only help understand what have impeded previous efforts on learning to fuse the multi-modal features, but also enhance the generalization ability of the learned fusion to novel test data. The new subclass based solution, with AP of 0.303 and P100 of 0.55 on the MediaEval 2015 test set, outperforms several state-of-the-art alternatives. Notice that our solution does not require fine-grained annotations on the test set, so it can be directly applied on novel and fully unlabeled videos. Interestingly, our study shows that motion related features, though being essential part in previous systems, are dispensable.
Tasks
Published	2016-04-27
URL	http://arxiv.org/abs/1604.08088v1
PDF	http://arxiv.org/pdf/1604.08088v1.pdf
PWC	https://paperswithcode.com/paper/detecting-violence-in-video-using-subclasses
Repo
Framework

An Approach to Stable Gradient Descent Adaptation of Higher-Order Neural Units


Title	An Approach to Stable Gradient Descent Adaptation of Higher-Order Neural Units
Authors	Ivo Bukovsky, Noriyasu Homma
Abstract	Stability evaluation of a weight-update system of higher-order neural units (HONUs) with polynomial aggregation of neural inputs (also known as classes of polynomial neural networks) for adaptation of both feedforward and recurrent HONUs by a gradient descent method is introduced. An essential core of the approach is based on spectral radius of a weight-update system, and it allows stability monitoring and its maintenance at every adaptation step individually. Assuring stability of the weight-update system (at every single adaptation step) naturally results in adaptation stability of the whole neural architecture that adapts to target data. As an aside, the used approach highlights the fact that the weight optimization of HONU is a linear problem, so the proposed approach can be generally extended to any neural architecture that is linear in its adaptable parameters.
Tasks
Published	2016-06-23
URL	http://arxiv.org/abs/1606.07149v1
PDF	http://arxiv.org/pdf/1606.07149v1.pdf
PWC	https://paperswithcode.com/paper/an-approach-to-stable-gradient-descent
Repo
Framework

Wikiometrics: A Wikipedia Based Ranking System


Title	Wikiometrics: A Wikipedia Based Ranking System
Authors	Gilad Katz, Lior Rokach
Abstract	We present a new concept - Wikiometrics - the derivation of metrics and indicators from Wikipedia. Wikipedia provides an accurate representation of the real world due to its size, structure, editing policy and popularity. We demonstrate an innovative mining methodology, where different elements of Wikipedia - content, structure, editorial actions and reader reviews - are used to rank items in a manner which is by no means inferior to rankings produced by experts or other methods. We test our proposed method by applying it to two real-world ranking problems: top world universities and academic journals. Our proposed ranking methods were compared to leading and widely accepted benchmarks, and were found to be extremely correlative but with the advantage of the data being publically available.
Tasks
Published	2016-01-06
URL	http://arxiv.org/abs/1601.01058v2
PDF	http://arxiv.org/pdf/1601.01058v2.pdf
PWC	https://paperswithcode.com/paper/wikiometrics-a-wikipedia-based-ranking-system
Repo
Framework

Asking the metaquestions in constraint tractability


Title	Asking the metaquestions in constraint tractability
Authors	Hubie Chen, Benoit Larose
Abstract	The constraint satisfaction problem (CSP) involves deciding, given a set of variables and a set of constraints on the variables, whether or not there is an assignment to the variables satisfying all of the constraints. One formulation of the CSP is as the problem of deciding, given a pair (G,H) of relational structures, whether or not there is a homomorphism from the first structure to the second structure. The CSP is in general NP-hard; a common way to restrict this problem is to fix the second structure H, so that each structure H gives rise to a problem CSP(H). The problem family CSP(H) has been studied using an algebraic approach, which links the algorithmic and complexity properties of each problem CSP(H) to a set of operations, the so-called polymorphisms of H. Certain types of polymorphisms are known to imply the polynomial-time tractability of $CSP(H)$, and others are conjectured to do so. This article systematically studies—for various classes of polymorphisms—the computational complexity of deciding whether or not a given structure H admits a polymorphism from the class. Among other results, we prove the NP-completeness of deciding a condition conjectured to characterize the tractable problems CSP(H), as well as the NP-completeness of deciding if CSP(H) has bounded width.
Tasks
Published	2016-04-04
URL	http://arxiv.org/abs/1604.00932v2
PDF	http://arxiv.org/pdf/1604.00932v2.pdf
PWC	https://paperswithcode.com/paper/asking-the-metaquestions-in-constraint
Repo
Framework

Scaled stochastic gradient descent for low-rank matrix completion


Title	Scaled stochastic gradient descent for low-rank matrix completion
Authors	Bamdev Mishra, Rodolphe Sepulchre
Abstract	The paper looks at a scaled variant of the stochastic gradient descent algorithm for the matrix completion problem. Specifically, we propose a novel matrix-scaling of the partial derivatives that acts as an efficient preconditioning for the standard stochastic gradient descent algorithm. This proposed matrix-scaling provides a trade-off between local and global second order information. It also resolves the issue of scale invariance that exists in matrix factorization models. The overall computational complexity is linear with the number of known entries, thereby extending to a large-scale setup. Numerical comparisons show that the proposed algorithm competes favorably with state-of-the-art algorithms on various different benchmarks.
Tasks	Low-Rank Matrix Completion, Matrix Completion
Published	2016-03-16
URL	http://arxiv.org/abs/1603.04989v2
PDF	http://arxiv.org/pdf/1603.04989v2.pdf
PWC	https://paperswithcode.com/paper/scaled-stochastic-gradient-descent-for-low
Repo
Framework

Permutation NMF


Title	Permutation NMF
Authors	Giovanni Barbarino
Abstract	Nonnegative Matrix Factorization(NMF) is a common used technique in machine learning to extract features out of data such as text documents and images thanks to its natural clustering properties. In particular, it is popular in image processing since it can decompose several pictures and recognize common parts if they’re located in the same position over the photos. This paper’s aim is to present a way to add the translation invariance to the classical NMF, that is, the algorithms presented are able to detect common features, even when they’re shifted, in different original images.
Tasks
Published	2016-08-03
URL	http://arxiv.org/abs/1608.01372v1
PDF	http://arxiv.org/pdf/1608.01372v1.pdf
PWC	https://paperswithcode.com/paper/permutation-nmf
Repo
Framework

M2CAI Workflow Challenge: Convolutional Neural Networks with Time Smoothing and Hidden Markov Model for Video Frames Classification


Title	M2CAI Workflow Challenge: Convolutional Neural Networks with Time Smoothing and Hidden Markov Model for Video Frames Classification
Authors	Rémi Cadène, Thomas Robert, Nicolas Thome, Matthieu Cord
Abstract	Our approach is among the three best to tackle the M2CAI Workflow challenge. The latter consists in recognizing the operation phase for each frames of endoscopic videos. In this technical report, we compare several classification models and temporal smoothing methods. Our submitted solution is a fine tuned Residual Network-200 on 80% of the training set with temporal smoothing using simple temporal averaging of the predictions and a Hidden Markov Model modeling the sequence.
Tasks
Published	2016-10-18
URL	http://arxiv.org/abs/1610.05541v2
PDF	http://arxiv.org/pdf/1610.05541v2.pdf
PWC	https://paperswithcode.com/paper/m2cai-workflow-challenge-convolutional-neural
Repo
Framework

A Repeated Signal Difference for Recognising Patterns


Title	A Repeated Signal Difference for Recognising Patterns
Authors	Kieran Greer
Abstract	This paper describes a new mechanism that might help with defining pattern sequences, by the fact that it can produce an upper bound on the ensemble value that can persistently oscillate with the actual values produced from each pattern. With every firing event, a node also receives an on/off feedback switch. If the node fires, then it sends a feedback result depending on the input signal strength. If the input signal is positive or larger, it can store an ‘on’ switch feedback for the next iteration. If the signal is negative or smaller, it can store an ‘off’ switch feedback for the next iteration. If the node does not fire, then it does not affect the current feedback situation and receives the switch command produced by the last active pattern event for the same neuron. The upper bound therefore also represents the largest or most enclosing pattern set and the lower value is for the actual set of firing patterns. If the pattern sequence repeats, it will oscillate between the two values, allowing them to be recognised and measured more easily, over time. Tests show that changing the sequence ordering produces different value sets, which can also be measured.
Tasks
Published	2016-04-18
URL	http://arxiv.org/abs/1604.05170v3
PDF	http://arxiv.org/pdf/1604.05170v3.pdf
PWC	https://paperswithcode.com/paper/a-repeated-signal-difference-for-recognising
Repo
Framework


Title	Forecasting Social Navigation in Crowded Complex Scenes
Authors	Alexandre Robicquet, Alexandre Alahi, Amir Sadeghian, Bryan Anenberg, John Doherty, Eli Wu, Silvio Savarese
Abstract	When humans navigate a crowed space such as a university campus or the sidewalks of a busy street, they follow common sense rules based on social etiquette. In this paper, we argue that in order to enable the design of new algorithms that can take fully advantage of these rules to better solve tasks such as target tracking or trajectory forecasting, we need to have access to better data in the first place. To that end, we contribute the very first large scale dataset (to the best of our knowledge) that collects images and videos of various types of targets (not just pedestrians, but also bikers, skateboarders, cars, buses, golf carts) that navigate in a real-world outdoor environment such as a university campus. We present an extensive evaluation where different methods for trajectory forecasting are evaluated and compared. Moreover, we present a new algorithm for trajectory prediction that exploits the complexity of our new dataset and allows to: i) incorporate inter-class interactions into trajectory prediction models (e.g, pedestrian vs bike) as opposed to just intra-class interactions (e.g., pedestrian vs pedestrian); ii) model the degree to which the social forces are regulating an interaction. We call the latter “social sensitivity"and it captures the sensitivity to which a target is responding to a certain interaction. An extensive experimental evaluation demonstrates the effectiveness of our novel approach.
Tasks	Common Sense Reasoning, Trajectory Prediction
Published	2016-01-05
URL	http://arxiv.org/abs/1601.00998v1
PDF	http://arxiv.org/pdf/1601.00998v1.pdf
PWC	https://paperswithcode.com/paper/forecasting-social-navigation-in-crowded
Repo
Framework

One-vs-Each Approximation to Softmax for Scalable Estimation of Probabilities


Title	One-vs-Each Approximation to Softmax for Scalable Estimation of Probabilities
Authors	Michalis K. Titsias
Abstract	The softmax representation of probabilities for categorical variables plays a prominent role in modern machine learning with numerous applications in areas such as large scale classification, neural language modeling and recommendation systems. However, softmax estimation is very expensive for large scale inference because of the high cost associated with computing the normalizing constant. Here, we introduce an efficient approximation to softmax probabilities which takes the form of a rigorous lower bound on the exact probability. This bound is expressed as a product over pairwise probabilities and it leads to scalable estimation based on stochastic optimization. It allows us to perform doubly stochastic estimation by subsampling both training instances and class labels. We show that the new bound has interesting theoretical properties and we demonstrate its use in classification problems.
Tasks	Language Modelling, Recommendation Systems, Stochastic Optimization
Published	2016-09-23
URL	http://arxiv.org/abs/1609.07410v2
PDF	http://arxiv.org/pdf/1609.07410v2.pdf
PWC	https://paperswithcode.com/paper/one-vs-each-approximation-to-softmax-for-1
Repo
Framework

A Spatial and Temporal Non-Local Filter Based Data Fusion


Title	A Spatial and Temporal Non-Local Filter Based Data Fusion
Authors	Qing Cheng, Huiqing Liu, Huanfeng Shen, Penghai Wu, Liangpei Zhang
Abstract	The trade-off in remote sensing instruments that balances the spatial resolution and temporal frequency limits our capacity to monitor spatial and temporal dynamics effectively. The spatiotemporal data fusion technique is considered as a cost-effective way to obtain remote sensing data with both high spatial resolution and high temporal frequency, by blending observations from multiple sensors with different advantages or characteristics. In this paper, we develop the spatial and temporal non-local filter based fusion model (STNLFFM) to enhance the prediction capacity and accuracy, especially for complex changed landscapes. The STNLFFM method provides a new transformation relationship between the fine-resolution reflectance images acquired from the same sensor at different dates with the help of coarse-resolution reflectance data, and makes full use of the high degree of spatiotemporal redundancy in the remote sensing image sequence to produce the final prediction. The proposed method was tested over both the Coleambally Irrigation Area study site and the Lower Gwydir Catchment study site. The results show that the proposed method can provide a more accurate and robust prediction, especially for heterogeneous landscapes and temporally dynamic areas.
Tasks
Published	2016-11-22
URL	http://arxiv.org/abs/1611.07231v1
PDF	http://arxiv.org/pdf/1611.07231v1.pdf
PWC	https://paperswithcode.com/paper/a-spatial-and-temporal-non-local-filter-based
Repo
Framework

Joint Copying and Restricted Generation for Paraphrase


Title	Joint Copying and Restricted Generation for Paraphrase
Authors	Ziqiang Cao, Chuwei Luo, Wenjie Li, Sujian Li
Abstract	Many natural language generation tasks, such as abstractive summarization and text simplification, are paraphrase-orientated. In these tasks, copying and rewriting are two main writing modes. Most previous sequence-to-sequence (Seq2Seq) models use a single decoder and neglect this fact. In this paper, we develop a novel Seq2Seq model to fuse a copying decoder and a restricted generative decoder. The copying decoder finds the position to be copied based on a typical attention model. The generative decoder produces words limited in the source-specific vocabulary. To combine the two decoders and determine the final output, we develop a predictor to predict the mode of copying or rewriting. This predictor can be guided by the actual writing mode in the training data. We conduct extensive experiments on two different paraphrase datasets. The result shows that our model outperforms the state-of-the-art approaches in terms of both informativeness and language quality.
Tasks	Abstractive Text Summarization, Text Generation, Text Simplification
Published	2016-11-28
URL	http://arxiv.org/abs/1611.09235v1
PDF	http://arxiv.org/pdf/1611.09235v1.pdf
PWC	https://paperswithcode.com/paper/joint-copying-and-restricted-generation-for
Repo
Framework

Imposing higher-level Structure in Polyphonic Music Generation using Convolutional Restricted Boltzmann Machines and Constraints


Title	Imposing higher-level Structure in Polyphonic Music Generation using Convolutional Restricted Boltzmann Machines and Constraints
Authors	Stefan Lattner, Maarten Grachten, Gerhard Widmer
Abstract	We introduce a method for imposing higher-level structure on generated, polyphonic music. A Convolutional Restricted Boltzmann Machine (C-RBM) as a generative model is combined with gradient descent constraint optimisation to provide further control over the generation process. Among other things, this allows for the use of a “template” piece, from which some structural properties can be extracted, and transferred as constraints to the newly generated material. The sampling process is guided with Simulated Annealing to avoid local optima, and to find solutions that both satisfy the constraints, and are relatively stable with respect to the C-RBM. Results show that with this approach it is possible to control the higher-level self-similarity structure, the meter, and the tonal properties of the resulting musical piece, while preserving its local musical coherence.
Tasks	Music Generation
Published	2016-12-14
URL	http://arxiv.org/abs/1612.04742v4
PDF	http://arxiv.org/pdf/1612.04742v4.pdf
PWC	https://paperswithcode.com/paper/imposing-higher-level-structure-in-polyphonic
Repo
Framework

Adaptive Newton Method for Empirical Risk Minimization to Statistical Accuracy


Title	Adaptive Newton Method for Empirical Risk Minimization to Statistical Accuracy
Authors	Aryan Mokhtari, Alejandro Ribeiro
Abstract	We consider empirical risk minimization for large-scale datasets. We introduce Ada Newton as an adaptive algorithm that uses Newton’s method with adaptive sample sizes. The main idea of Ada Newton is to increase the size of the training set by a factor larger than one in a way that the minimization variable for the current training set is in the local neighborhood of the optimal argument of the next training set. This allows to exploit the quadratic convergence property of Newton’s method and reach the statistical accuracy of each training set with only one iteration of Newton’s method. We show theoretically and empirically that Ada Newton can double the size of the training set in each iteration to achieve the statistical accuracy of the full training set with about two passes over the dataset.
Tasks
Published	2016-05-24
URL	http://arxiv.org/abs/1605.07659v1
PDF	http://arxiv.org/pdf/1605.07659v1.pdf
PWC	https://paperswithcode.com/paper/adaptive-newton-method-for-empirical-risk
Repo
Framework

Extend natural neighbor: a novel classification method with self-adaptive neighborhood parameters in different stages


Title	Extend natural neighbor: a novel classification method with self-adaptive neighborhood parameters in different stages
Authors	Ji Feng, Qingsheng Zhu, Jinlong Huang, Lijun Yang
Abstract	Various kinds of k-nearest neighbor (KNN) based classification methods are the bases of many well-established and high-performance pattern-recognition techniques, but both of them are vulnerable to their parameter choice. Essentially, the challenge is to detect the neighborhood of various data sets, while utterly ignorant of the data characteristic. This article introduces a new supervised classification method: the extend natural neighbor (ENaN) method, and shows that it provides a better classification result without choosing the neighborhood parameter artificially. Unlike the original KNN based method which needs a prior k, the ENaNE method predicts different k in different stages. Therefore, the ENaNE method is able to learn more from flexible neighbor information both in training stage and testing stage, and provide a better classification result.
Tasks
Published	2016-12-07
URL	http://arxiv.org/abs/1612.02310v1
PDF	http://arxiv.org/pdf/1612.02310v1.pdf
PWC	https://paperswithcode.com/paper/extend-natural-neighbor-a-novel
Repo
Framework