May 6, 2019

2799 words 14 mins read

Paper Group ANR 333

Learning Spatially Regularized Correlation Filters for Visual Tracking. Track Facial Points in Unconstrained Videos. Pilot Testing an Artificial Intelligence Algorithm That Selects Homeless Youth Peer Leaders Who Promote HIV Testing. Tensor Methods and Recommender Systems. Recurrent 3D Attentional Networks for End-to-End Active Object Recognition. …

Learning Spatially Regularized Correlation Filters for Visual Tracking


Title	Learning Spatially Regularized Correlation Filters for Visual Tracking
Authors	Martin Danelljan, Gustav Häger, Fahad Shahbaz Khan, Michael Felsberg
Abstract	Robust and accurate visual tracking is one of the most challenging computer vision problems. Due to the inherent lack of training data, a robust approach for constructing a target appearance model is crucial. Recently, discriminatively learned correlation filters (DCF) have been successfully applied to address this problem for tracking. These methods utilize a periodic assumption of the training samples to efficiently learn a classifier on all patches in the target neighborhood. However, the periodic assumption also introduces unwanted boundary effects, which severely degrade the quality of the tracking model. We propose Spatially Regularized Discriminative Correlation Filters (SRDCF) for tracking. A spatial regularization component is introduced in the learning to penalize correlation filter coefficients depending on their spatial location. Our SRDCF formulation allows the correlation filters to be learned on a significantly larger set of negative training samples, without corrupting the positive samples. We further propose an optimization strategy, based on the iterative Gauss-Seidel method, for efficient online learning of our SRDCF. Experiments are performed on four benchmark datasets: OTB-2013, ALOV++, OTB-2015, and VOT2014. Our approach achieves state-of-the-art results on all four datasets. On OTB-2013 and OTB-2015, we obtain an absolute gain of 8.0% and 8.2% respectively, in mean overlap precision, compared to the best existing trackers.
Tasks	Visual Tracking
Published	2016-08-19
URL	http://arxiv.org/abs/1608.05571v1
PDF	http://arxiv.org/pdf/1608.05571v1.pdf
PWC	https://paperswithcode.com/paper/learning-spatially-regularized-correlation
Repo
Framework

Track Facial Points in Unconstrained Videos


Title	Track Facial Points in Unconstrained Videos
Authors	Xi Peng, Qiong Hu, Junzhou Huang, Dimitris N. Metaxas
Abstract	Tracking Facial Points in unconstrained videos is challenging due to the non-rigid deformation that changes over time. In this paper, we propose to exploit incremental learning for person-specific alignment in wild conditions. Our approach takes advantage of part-based representation and cascade regression for robust and efficient alignment on each frame. Unlike existing methods that usually rely on models trained offline, we incrementally update the representation subspace and the cascade of regressors in a unified framework to achieve personalized modeling on the fly. To alleviate the drifting issue, the fitting results are evaluated using a deep neural network, where well-aligned faces are picked out to incrementally update the representation and fitting models. Both image and video datasets are employed to valid the proposed method. The results demonstrate the superior performance of our approach compared with existing approaches in terms of fitting accuracy and efficiency.
Tasks
Published	2016-09-09
URL	http://arxiv.org/abs/1609.02825v1
PDF	http://arxiv.org/pdf/1609.02825v1.pdf
PWC	https://paperswithcode.com/paper/track-facial-points-in-unconstrained-videos
Repo
Framework

Pilot Testing an Artificial Intelligence Algorithm That Selects Homeless Youth Peer Leaders Who Promote HIV Testing


Title	Pilot Testing an Artificial Intelligence Algorithm That Selects Homeless Youth Peer Leaders Who Promote HIV Testing
Authors	Eric Rice, Robin Petering, Jaih Craddock, Amanda Yoshioka-Maxwell, Amulya Yadav, Milind Tambe
Abstract	Objective. To pilot test an artificial intelligence (AI) algorithm that selects peer change agents (PCA) to disseminate HIV testing messaging in a population of homeless youth. Methods. We recruited and assessed 62 youth at baseline, 1 month (n = 48), and 3 months (n = 38). A Facebook app collected preliminary social network data. Eleven PCAs selected by AI attended a 1-day training and 7 weekly booster sessions. Mixed-effects models with random effects were used to assess change over time. Results. Significant change over time was observed in past 6-month HIV testing (57.9%, 82.4%, 76.3%; p < .05) but not condom use (63.9%, 65.7%, 65.8%). Most youth reported speaking to a PCA about HIV prevention (72.0% at 1 month, 61.5% at 3 months). Conclusions. AI is a promising avenue for implementing PCA models for homeless youth. Increasing rates of regular HIV testing is critical to HIV prevention and linking homeless youth to treatment.
Tasks
Published	2016-08-19
URL	http://arxiv.org/abs/1608.05701v1
PDF	http://arxiv.org/pdf/1608.05701v1.pdf
PWC	https://paperswithcode.com/paper/pilot-testing-an-artificial-intelligence
Repo
Framework

Tensor Methods and Recommender Systems


Title	Tensor Methods and Recommender Systems
Authors	Evgeny Frolov, Ivan Oseledets
Abstract	A substantial progress in development of new and efficient tensor factorization techniques has led to an extensive research of their applicability in recommender systems field. Tensor-based recommender models push the boundaries of traditional collaborative filtering techniques by taking into account a multifaceted nature of real environments, which allows to produce more accurate, situational (e.g. context-aware, criteria-driven) recommendations. Despite the promising results, tensor-based methods are poorly covered in existing recommender systems surveys. This survey aims to complement previous works and provide a comprehensive overview on the subject. To the best of our knowledge, this is the first attempt to consolidate studies from various application domains in an easily readable, digestible format, which helps to get a notion of the current state of the field. We also provide a high level discussion of the future perspectives and directions for further improvement of tensor-based recommendation systems.
Tasks	Recommendation Systems
Published	2016-03-19
URL	http://arxiv.org/abs/1603.06038v2
PDF	http://arxiv.org/pdf/1603.06038v2.pdf
PWC	https://paperswithcode.com/paper/tensor-methods-and-recommender-systems
Repo
Framework

Recurrent 3D Attentional Networks for End-to-End Active Object Recognition


Title	Recurrent 3D Attentional Networks for End-to-End Active Object Recognition
Authors	Min Liu, Yifei Shi, Lintao Zheng, Kai Xu, Hui Huang, Dinesh Manocha
Abstract	Active vision is inherently attention-driven: The agent actively selects views to attend in order to fast achieve the vision task while improving its internal representation of the scene being observed. Inspired by the recent success of attention-based models in 2D vision tasks based on single RGB images, we propose to address the multi-view depth-based active object recognition using attention mechanism, through developing an end-to-end recurrent 3D attentional network. The architecture takes advantage of a recurrent neural network (RNN) to store and update an internal representation. Our model, trained with 3D shape datasets, is able to iteratively attend to the best views targeting an object of interest for recognizing it. To realize 3D view selection, we derive a 3D spatial transformer network which is differentiable for training with backpropagation, achieving much faster convergence than the reinforcement learning employed by most existing attention-based models. Experiments show that our method, with only depth input, achieves state-of-the-art next-best-view performance in time efficiency and recognition accuracy.
Tasks	Object Recognition
Published	2016-10-14
URL	http://arxiv.org/abs/1610.04308v3
PDF	http://arxiv.org/pdf/1610.04308v3.pdf
PWC	https://paperswithcode.com/paper/recurrent-3d-attentional-networks-for-end-to
Repo
Framework

On Clustering Time Series Using Euclidean Distance and Pearson Correlation


Title	On Clustering Time Series Using Euclidean Distance and Pearson Correlation
Authors	Michael R. Berthold, Frank Höppner
Abstract	For time series comparisons, it has often been observed that z-score normalized Euclidean distances far outperform the unnormalized variant. In this paper we show that a z-score normalized, squared Euclidean Distance is, in fact, equal to a distance based on Pearson Correlation. This has profound impact on many distance-based classification or clustering methods. In addition to this theoretically sound result we also show that the often used k-Means algorithm formally needs a mod ification to keep the interpretation as Pearson correlation strictly valid. Experimental results demonstrate that in many cases the standard k-Means algorithm generally produces the same results.
Tasks	Time Series
Published	2016-01-10
URL	http://arxiv.org/abs/1601.02213v1
PDF	http://arxiv.org/pdf/1601.02213v1.pdf
PWC	https://paperswithcode.com/paper/on-clustering-time-series-using-euclidean
Repo
Framework

DeepSense: A Unified Deep Learning Framework for Time-Series Mobile Sensing Data Processing


Title	DeepSense: A Unified Deep Learning Framework for Time-Series Mobile Sensing Data Processing
Authors	Shuochao Yao, Shaohan Hu, Yiran Zhao, Aston Zhang, Tarek Abdelzaher
Abstract	Mobile sensing applications usually require time-series inputs from sensors. Some applications, such as tracking, can use sensed acceleration and rate of rotation to calculate displacement based on physical system models. Other applications, such as activity recognition, extract manually designed features from sensor inputs for classification. Such applications face two challenges. On one hand, on-device sensor measurements are noisy. For many mobile applications, it is hard to find a distribution that exactly describes the noise in practice. Unfortunately, calculating target quantities based on physical system and noise models is only as accurate as the noise assumptions. Similarly, in classification applications, although manually designed features have proven to be effective, it is not always straightforward to find the most robust features to accommodate diverse sensor noise patterns and user behaviors. To this end, we propose DeepSense, a deep learning framework that directly addresses the aforementioned noise and feature customization challenges in a unified manner. DeepSense integrates convolutional and recurrent neural networks to exploit local interactions among similar mobile sensors, merge local interactions of different sensory modalities into global interactions, and extract temporal relationships to model signal dynamics. DeepSense thus provides a general signal estimation and classification framework that accommodates a wide range of applications. We demonstrate the effectiveness of DeepSense using three representative and challenging tasks: car tracking with motion sensors, heterogeneous human activity recognition, and user identification with biometric motion analysis. DeepSense significantly outperforms the state-of-the-art methods for all three tasks. In addition, DeepSense is feasible to implement on smartphones due to its moderate energy consumption and low latency
Tasks	Activity Recognition, Human Activity Recognition, Time Series
Published	2016-11-07
URL	http://arxiv.org/abs/1611.01942v2
PDF	http://arxiv.org/pdf/1611.01942v2.pdf
PWC	https://paperswithcode.com/paper/deepsense-a-unified-deep-learning-framework
Repo
Framework

Training Dependency Parsers with Partial Annotation


Title	Training Dependency Parsers with Partial Annotation
Authors	Zhenghua Li, Yue Zhang, Jiayuan Chao, Min Zhang
Abstract	Recently, these has been a surge on studying how to obtain partially annotated data for model supervision. However, there still lacks a systematic study on how to train statistical models with partial annotation (PA). Taking dependency parsing as our case study, this paper describes and compares two straightforward approaches for three mainstream dependency parsers. The first approach is previously proposed to directly train a log-linear graph-based parser (LLGPar) with PA based on a forest-based objective. This work for the first time proposes the second approach to directly training a linear graph-based parse (LGPar) and a linear transition-based parser (LTPar) with PA based on the idea of constrained decoding. We conduct extensive experiments on Penn Treebank under three different settings for simulating PA, i.e., random dependencies, most uncertain dependencies, and dependencies with divergent outputs from the three parsers. The results show that LLGPar is most effective in learning from PA and LTPar lags behind the graph-based counterparts by large margin. Moreover, LGPar and LTPar can achieve best performance by using LLGPar to complete PA into full annotation (FA).
Tasks	Dependency Parsing
Published	2016-09-29
URL	http://arxiv.org/abs/1609.09247v1
PDF	http://arxiv.org/pdf/1609.09247v1.pdf
PWC	https://paperswithcode.com/paper/training-dependency-parsers-with-partial
Repo
Framework

Monte Carlo Tableau Proof Search


Title	Monte Carlo Tableau Proof Search
Authors	Michael Färber, Cezary Kaliszyk, Josef Urban
Abstract	We study Monte Carlo Tree Search to guide proof search in tableau calculi. This includes proposing a number of proof-state evaluation heuristics, some of which are learnt from previous proofs. We present an implementation based on the leanCoP prover. The system is trained and evaluated on a large suite of related problems coming from the Mizar proof assistant, showing that it is capable to find new and different proofs.
Tasks	Automated Theorem Proving
Published	2016-11-18
URL	https://arxiv.org/abs/1611.05990v2
PDF	https://arxiv.org/pdf/1611.05990v2.pdf
PWC	https://paperswithcode.com/paper/monte-carlo-connection-prover
Repo
Framework

Convexification of Learning from Constraints


Title	Convexification of Learning from Constraints
Authors	Iaroslav Shcherbatyi, Bjoern Andres
Abstract	Regularized empirical risk minimization with constrained labels (in contrast to fixed labels) is a remarkably general abstraction of learning. For common loss and regularization functions, this optimization problem assumes the form of a mixed integer program (MIP) whose objective function is non-convex. In this form, the problem is resistant to standard optimization techniques. We construct MIPs with the same solutions whose objective functions are convex. Specifically, we characterize the tightest convex extension of the objective function, given by the Legendre-Fenchel biconjugate. Computing values of this tightest convex extension is NP-hard. However, by applying our characterization to every function in an additive decomposition of the objective function, we obtain a class of looser convex extensions that can be computed efficiently. For some decompositions, common loss and regularization functions, we derive a closed form.
Tasks
Published	2016-02-22
URL	http://arxiv.org/abs/1602.06746v1
PDF	http://arxiv.org/pdf/1602.06746v1.pdf
PWC	https://paperswithcode.com/paper/convexification-of-learning-from-constraints
Repo
Framework

Training Input-Output Recurrent Neural Networks through Spectral Methods


Title	Training Input-Output Recurrent Neural Networks through Spectral Methods
Authors	Hanie Sedghi, Anima Anandkumar
Abstract	We consider the problem of training input-output recurrent neural networks (RNN) for sequence labeling tasks. We propose a novel spectral approach for learning the network parameters. It is based on decomposition of the cross-moment tensor between the output and a non-linear transformation of the input, based on score functions. We guarantee consistent learning with polynomial sample and computational complexity under transparent conditions such as non-degeneracy of model parameters, polynomial activations for the neurons, and a Markovian evolution of the input sequence. We also extend our results to Bidirectional RNN which uses both previous and future information to output the label at each time point, and is employed in many NLP tasks such as POS tagging.
Tasks
Published	2016-03-03
URL	http://arxiv.org/abs/1603.00954v5
PDF	http://arxiv.org/pdf/1603.00954v5.pdf
PWC	https://paperswithcode.com/paper/training-input-output-recurrent-neural
Repo
Framework

Theory and Tools for the Conversion of Analog to Spiking Convolutional Neural Networks


Title	Theory and Tools for the Conversion of Analog to Spiking Convolutional Neural Networks
Authors	Bodo Rueckauer, Iulia-Alexandra Lungu, Yuhuang Hu, Michael Pfeiffer
Abstract	Deep convolutional neural networks (CNNs) have shown great potential for numerous real-world machine learning applications, but performing inference in large CNNs in real-time remains a challenge. We have previously demonstrated that traditional CNNs can be converted into deep spiking neural networks (SNNs), which exhibit similar accuracy while reducing both latency and computational load as a consequence of their data-driven, event-based style of computing. Here we provide a novel theory that explains why this conversion is successful, and derive from it several new tools to convert a larger and more powerful class of deep networks into SNNs. We identify the main sources of approximation errors in previous conversion methods, and propose simple mechanisms to fix these issues. Furthermore, we develop spiking implementations of common CNN operations such as max-pooling, softmax, and batch-normalization, which allow almost loss-less conversion of arbitrary CNN architectures into the spiking domain. Empirical evaluation of different network architectures on the MNIST and CIFAR10 benchmarks leads to the best SNN results reported to date.
Tasks
Published	2016-12-13
URL	http://arxiv.org/abs/1612.04052v1
PDF	http://arxiv.org/pdf/1612.04052v1.pdf
PWC	https://paperswithcode.com/paper/theory-and-tools-for-the-conversion-of-analog
Repo
Framework

A global constraint for closed itemset mining


Title	A global constraint for closed itemset mining
Authors	Mehdi Maamar, Nadjib Lazaar, Samir Loudni, Yahia Lebbah
Abstract	Discovering the set of closed frequent patterns is one of the fundamental problems in Data Mining. Recent Constraint Programming (CP) approaches for declarative itemset mining have proven their usefulness and flexibility. But the wide use of reified constraints in current CP approaches raises many difficulties to cope with high dimensional datasets. This paper proposes CLOSED PATTERN global constraint which does not require any reified constraints nor any extra variables to encode efficiently the Closed Frequent Pattern Mining (CFPM) constraint. CLOSED-PATTERN captures the particular semantics of the CFPM problem in order to ensure a polynomial pruning algorithm ensuring domain consistency. The computational properties of our constraint are analyzed and their practical effectiveness is experimentally evaluated.
Tasks
Published	2016-04-17
URL	http://arxiv.org/abs/1604.04894v1
PDF	http://arxiv.org/pdf/1604.04894v1.pdf
PWC	https://paperswithcode.com/paper/a-global-constraint-for-closed-itemset-mining
Repo
Framework

Information-Theoretic Bounds and Approximations in Neural Population Coding


Title	Information-Theoretic Bounds and Approximations in Neural Population Coding
Authors	Wentao Huang, Kechen Zhang
Abstract	While Shannon’s mutual information has widespread applications in many disciplines, for practical applications it is often difficult to calculate its value accurately for high-dimensional variables because of the curse of dimensionality. This paper is focused on effective approximation methods for evaluating mutual information in the context of neural population coding. For large but finite neural populations, we derive several information-theoretic asymptotic bounds and approximation formulas that remain valid in high-dimensional spaces. We prove that optimizing the population density distribution based on these approximation formulas is a convex optimization problem which allows efficient numerical solutions. Numerical simulation results confirmed that our asymptotic formulas were highly accurate for approximating mutual information for large neural populations. In special cases, the approximation formulas are exactly equal to the true mutual information. We also discuss techniques of variable transformation and dimensionality reduction to facilitate computation of the approximations.
Tasks	Dimensionality Reduction
Published	2016-11-04
URL	http://arxiv.org/abs/1611.01414v3
PDF	http://arxiv.org/pdf/1611.01414v3.pdf
PWC	https://paperswithcode.com/paper/information-theoretic-bounds-and
Repo
Framework

Imitation Learning with Recurrent Neural Networks


Title	Imitation Learning with Recurrent Neural Networks
Authors	Khanh Nguyen
Abstract	We present a novel view that unifies two frameworks that aim to solve sequential prediction problems: learning to search (L2S) and recurrent neural networks (RNN). We point out equivalences between elements of the two frameworks. By complementing what is missing from one framework comparing to the other, we introduce a more advanced imitation learning framework that, on one hand, augments L2S s notion of search space and, on the other hand, enhances RNNs training procedure to be more robust to compounding errors arising from training on highly correlated examples.
Tasks	Imitation Learning
Published	2016-07-18
URL	http://arxiv.org/abs/1607.05241v1
PDF	http://arxiv.org/pdf/1607.05241v1.pdf
PWC	https://paperswithcode.com/paper/imitation-learning-with-recurrent-neural
Repo
Framework