May 6, 2019

2923 words 14 mins read

Paper Group ANR 388

Advancing Bayesian Optimization: The Mixed-Global-Local (MGL) Kernel and Length-Scale Cool Down. Quickest Moving Object Detection. Nonparametric Bayesian Topic Modelling with the Hierarchical Pitman-Yor Processes. An Oracle Inequality for Quasi-Bayesian Non-Negative Matrix Factorization. CGMOS: Certainty Guided Minority OverSampling. An Interactive …

Advancing Bayesian Optimization: The Mixed-Global-Local (MGL) Kernel and Length-Scale Cool Down


Title	Advancing Bayesian Optimization: The Mixed-Global-Local (MGL) Kernel and Length-Scale Cool Down
Authors	Kim Peter Wabersich, Marc Toussaint
Abstract	Bayesian Optimization (BO) has become a core method for solving expensive black-box optimization problems. While much research focussed on the choice of the acquisition function, we focus on online length-scale adaption and the choice of kernel function. Instead of choosing hyperparameters in view of maximum likelihood on past data, we propose to use the acquisition function to decide on hyperparameter adaptation more robustly and in view of the future optimization progress. Further, we propose a particular kernel function that includes non-stationarity and local anisotropy and thereby implicitly integrates the efficiency of local convex optimization with global Bayesian optimization. Comparisons to state-of-the art BO methods underline the efficiency of these mechanisms on global optimization benchmarks.
Tasks
Published	2016-12-09
URL	http://arxiv.org/abs/1612.03117v1
PDF	http://arxiv.org/pdf/1612.03117v1.pdf
PWC	https://paperswithcode.com/paper/advancing-bayesian-optimization-the-mixed
Repo
Framework

Quickest Moving Object Detection


Title	Quickest Moving Object Detection
Authors	Dong Lao, Ganesh Sundaramoorthi
Abstract	We present a general framework and method for simultaneous detection and segmentation of an object in a video that moves (or comes into view of the camera) at some unknown time in the video. The method is an online approach based on motion segmentation, and it operates under dynamic backgrounds caused by a moving camera or moving nuisances. The goal of the method is to detect and segment the object as soon as it moves. Due to stochastic variability in the video and unreliability of the motion signal, several frames are needed to reliably detect the object. The method is designed to detect and segment with minimum delay subject to a constraint on the false alarm rate. The method is derived as a problem of Quickest Change Detection. Experiments on a dataset show the effectiveness of our method in minimizing detection delay subject to false alarm constraints.
Tasks	Motion Segmentation, Object Detection
Published	2016-05-24
URL	http://arxiv.org/abs/1605.07369v1
PDF	http://arxiv.org/pdf/1605.07369v1.pdf
PWC	https://paperswithcode.com/paper/quickest-moving-object-detection
Repo
Framework

Nonparametric Bayesian Topic Modelling with the Hierarchical Pitman-Yor Processes


Title	Nonparametric Bayesian Topic Modelling with the Hierarchical Pitman-Yor Processes
Authors	Kar Wai Lim, Wray Buntine, Changyou Chen, Lan Du
Abstract	The Dirichlet process and its extension, the Pitman-Yor process, are stochastic processes that take probability distributions as a parameter. These processes can be stacked up to form a hierarchical nonparametric Bayesian model. In this article, we present efficient methods for the use of these processes in this hierarchical context, and apply them to latent variable models for text analytics. In particular, we propose a general framework for designing these Bayesian models, which are called topic models in the computer science community. We then propose a specific nonparametric Bayesian topic model for modelling text from social media. We focus on tweets (posts on Twitter) in this article due to their ease of access. We find that our nonparametric model performs better than existing parametric models in both goodness of fit and real world applications.
Tasks	Latent Variable Models, Topic Models
Published	2016-09-22
URL	http://arxiv.org/abs/1609.06783v1
PDF	http://arxiv.org/pdf/1609.06783v1.pdf
PWC	https://paperswithcode.com/paper/nonparametric-bayesian-topic-modelling-with
Repo
Framework

An Oracle Inequality for Quasi-Bayesian Non-Negative Matrix Factorization


Title	An Oracle Inequality for Quasi-Bayesian Non-Negative Matrix Factorization
Authors	Pierre Alquier, Benjamin Guedj
Abstract	The aim of this paper is to provide some theoretical understanding of quasi-Bayesian aggregation methods non-negative matrix factorization. We derive an oracle inequality for an aggregated estimator. This result holds for a very general class of prior distributions and shows how the prior affects the rate of convergence.
Tasks
Published	2016-01-06
URL	http://arxiv.org/abs/1601.01345v4
PDF	http://arxiv.org/pdf/1601.01345v4.pdf
PWC	https://paperswithcode.com/paper/an-oracle-inequality-for-quasi-bayesian-non
Repo
Framework

CGMOS: Certainty Guided Minority OverSampling


Title	CGMOS: Certainty Guided Minority OverSampling
Authors	Xi Zhang, Di Ma, Lin Gan, Shanshan Jiang, Gady Agam
Abstract	Handling imbalanced datasets is a challenging problem that if not treated correctly results in reduced classification performance. Imbalanced datasets are commonly handled using minority oversampling, whereas the SMOTE algorithm is a successful oversampling algorithm with numerous extensions. SMOTE extensions do not have a theoretical guarantee during training to work better than SMOTE and in many instances their performance is data dependent. In this paper we propose a novel extension to the SMOTE algorithm with a theoretical guarantee for improved classification performance. The proposed approach considers the classification performance of both the majority and minority classes. In the proposed approach CGMOS (Certainty Guided Minority OverSampling) new data points are added by considering certainty changes in the dataset. The paper provides a proof that the proposed algorithm is guaranteed to work better than SMOTE for training data. Further experimental results on 30 real-world datasets show that CGMOS works better than existing algorithms when using 6 different classifiers.
Tasks
Published	2016-07-21
URL	http://arxiv.org/abs/1607.06525v1
PDF	http://arxiv.org/pdf/1607.06525v1.pdf
PWC	https://paperswithcode.com/paper/cgmos-certainty-guided-minority-oversampling
Repo
Framework

An Interactive Machine Learning Framework


Title	An Interactive Machine Learning Framework
Authors	Teng Lee, James Johnson, Steve Cheng
Abstract	Machine learning (ML) is believed to be an effective and efficient tool to build reliable prediction model or extract useful structure from an avalanche of data. However, ML is also criticized by its difficulty in interpretation and complicated parameter tuning. In contrast, visualization is able to well organize and visually encode the entangled information in data and guild audiences to simpler perceptual inferences and analytic thinking. But large scale and high dimensional data will usually lead to the failure of many visualization methods. In this paper, we close a loop between ML and visualization via interaction between ML algorithm and users, so machine intelligence and human intelligence can cooperate and improve each other in a mutually rewarding way. In particular, we propose “transparent boosting tree (TBT)", which visualizes both the model structure and prediction statistics of each step in the learning process of gradient boosting tree to user, and involves user’s feedback operations to trees into the learning process. In TBT, ML is in charge of updating weights in learning model and filtering information shown to user from the big data, while visualization is in charge of providing a visual understanding of ML model to facilitate user exploration. It combines the advantages of both ML in big data statistics and human in decision making based on domain knowledge. We develop a user friendly interface for this novel learning method, and apply it to two datasets collected from real applications. Our study shows that making ML transparent by using interactive visualization can significantly improve the exploration of ML algorithms, give rise to novel insights of ML models, and integrates both machine and human intelligence.
Tasks	Decision Making
Published	2016-10-18
URL	http://arxiv.org/abs/1610.05463v1
PDF	http://arxiv.org/pdf/1610.05463v1.pdf
PWC	https://paperswithcode.com/paper/an-interactive-machine-learning-framework
Repo
Framework

Open challenges in understanding development and evolution of speech forms: The roles of embodied self-organization, motivation and active exploration


Title	Open challenges in understanding development and evolution of speech forms: The roles of embodied self-organization, motivation and active exploration
Authors	Pierre-Yves Oudeyer
Abstract	This article discusses open scientific challenges for understanding development and evolution of speech forms, as a commentary to Moulin-Frier et al. (Moulin-Frier et al., 2015). Based on the analysis of mathematical models of the origins of speech forms, with a focus on their assumptions , we study the fundamental question of how speech can be formed out of non–speech, at both developmental and evolutionary scales. In particular, we emphasize the importance of embodied self-organization , as well as the role of mechanisms of motivation and active curiosity-driven exploration in speech formation. Finally , we discuss an evolutionary-developmental perspective of the origins of speech.
Tasks
Published	2016-01-05
URL	http://arxiv.org/abs/1601.00816v1
PDF	http://arxiv.org/pdf/1601.00816v1.pdf
PWC	https://paperswithcode.com/paper/open-challenges-in-understanding-development
Repo
Framework

Tensor Decompositions via Two-Mode Higher-Order SVD (HOSVD)


Title	Tensor Decompositions via Two-Mode Higher-Order SVD (HOSVD)
Authors	Miaoyan Wang, Yun S. Song
Abstract	Tensor decompositions have rich applications in statistics and machine learning, and developing efficient, accurate algorithms for the problem has received much attention recently. Here, we present a new method built on Kruskal’s uniqueness theorem to decompose symmetric, nearly orthogonally decomposable tensors. Unlike the classical higher-order singular value decomposition which unfolds a tensor along a single mode, we consider unfoldings along two modes and use rank-1 constraints to characterize the underlying components. This tensor decomposition method provably handles a greater level of noise compared to previous methods and achieves a high estimation accuracy. Numerical results demonstrate that our algorithm is robust to various noise distributions and that it performs especially favorably as the order increases.
Tasks
Published	2016-12-12
URL	http://arxiv.org/abs/1612.03839v2
PDF	http://arxiv.org/pdf/1612.03839v2.pdf
PWC	https://paperswithcode.com/paper/tensor-decompositions-via-two-mode-higher
Repo
Framework

Object Classification with Joint Projection and Low-rank Dictionary Learning


Title	Object Classification with Joint Projection and Low-rank Dictionary Learning
Authors	Homa Foroughi, Nilanjan Ray, Hong Zhang
Abstract	For an object classification system, the most critical obstacles towards real-world applications are often caused by large intra-class variability, arising from different lightings, occlusion and corruption, in limited sample sets. Most methods in the literature would fail when the training samples are heavily occluded, corrupted or have significant illumination or viewpoint variations. Besides, most of the existing methods and especially deep learning-based methods, need large training sets to achieve a satisfactory recognition performance. Although using the pre-trained network on a generic large-scale dataset and fine-tune it to the small-sized target dataset is a widely used technique, this would not help when the content of base and target datasets are very different. To address these issues, we propose a joint projection and low-rank dictionary learning method using dual graph constraints (JP-LRDL). The proposed joint learning method would enable us to learn the features on top of which dictionaries can be better learned, from the data with large intra-class variability. Specifically, a structured class-specific dictionary is learned and the discrimination is further improved by imposing a graph constraint on the coding coefficients, that maximizes the intra-class compactness and inter-class separability. We also enforce low-rank and structural incoherence constraints on sub-dictionaries to make them more compact and robust to variations and outliers and reduce the redundancy among them, respectively. To preserve the intrinsic structure of data and penalize unfavourable relationship among training samples simultaneously, we introduce a projection graph into the framework, which significantly enhances the discriminative ability of the projection matrix and makes the method robust to small-sized and high-dimensional datasets.
Tasks	Dictionary Learning, Object Classification
Published	2016-12-05
URL	http://arxiv.org/abs/1612.01594v1
PDF	http://arxiv.org/pdf/1612.01594v1.pdf
PWC	https://paperswithcode.com/paper/object-classification-with-joint-projection
Repo
Framework

Volumetric Light-field Encryption at the Microscopic Scale


Title	Volumetric Light-field Encryption at the Microscopic Scale
Authors	Haoyu Li, Changliang Guo, Inbarasan Muniraj, Bryce C. Schroeder, John T. Sheridan, Shu Jia
Abstract	We report a light-field based method that allows the optical encryption of three-dimensional (3D) volumetric information at the microscopic scale in a single 2D light-field image. The system consists of a microlens array and an array of random phase/amplitude masks. The method utilizes a wave optics model to account for the dominant diffraction effect at this new scale, and the system point-spread function (PSF) serves as the key for encryption and decryption. We successfully developed and demonstrated a deconvolution algorithm to retrieve spatially multiplexed discrete and continuous volumetric data from 2D light-field images. Showing that the method is practical for data transmission and storage, we obtained a faithful reconstruction of the 3D volumetric information from a digital copy of the encrypted light-field image. The method represents a new level of optical encryption, paving the way for broad industrial and biomedical applications in processing and securing 3D data at the microscopic scale.
Tasks
Published	2016-10-26
URL	http://arxiv.org/abs/1610.08762v1
PDF	http://arxiv.org/pdf/1610.08762v1.pdf
PWC	https://paperswithcode.com/paper/volumetric-light-field-encryption-at-the
Repo
Framework

Constraint Selection in Metric Learning


Title	Constraint Selection in Metric Learning
Authors	Hoel Le Capitaine
Abstract	A number of machine learning algorithms are using a metric, or a distance, in order to compare individuals. The Euclidean distance is usually employed, but it may be more efficient to learn a parametric distance such as Mahalanobis metric. Learning such a metric is a hot topic since more than ten years now, and a number of methods have been proposed to efficiently learn it. However, the nature of the problem makes it quite difficult for large scale data, as well as data for which classes overlap. This paper presents a simple way of improving accuracy and scalability of any iterative metric learning algorithm, where constraints are obtained prior to the algorithm. The proposed approach relies on a loss-dependent weighted selection of constraints that are used for learning the metric. Using the corresponding dedicated loss function, the method clearly allows to obtain better results than state-of-the-art methods, both in terms of accuracy and time complexity. Some experimental results on real world, and potentially large, datasets are demonstrating the effectiveness of our proposition.
Tasks	Metric Learning
Published	2016-12-14
URL	http://arxiv.org/abs/1612.04853v1
PDF	http://arxiv.org/pdf/1612.04853v1.pdf
PWC	https://paperswithcode.com/paper/constraint-selection-in-metric-learning
Repo
Framework

Real-Time Hand Tracking Using a Sum of Anisotropic Gaussians Model


Title	Real-Time Hand Tracking Using a Sum of Anisotropic Gaussians Model
Authors	Srinath Sridhar, Helge Rhodin, Hans-Peter Seidel, Antti Oulasvirta, Christian Theobalt
Abstract	Real-time marker-less hand tracking is of increasing importance in human-computer interaction. Robust and accurate tracking of arbitrary hand motion is a challenging problem due to the many degrees of freedom, frequent self-occlusions, fast motions, and uniform skin color. In this paper, we propose a new approach that tracks the full skeleton motion of the hand from multiple RGB cameras in real-time. The main contributions include a new generative tracking method which employs an implicit hand shape representation based on Sum of Anisotropic Gaussians (SAG), and a pose fitting energy that is smooth and analytically differentiable making fast gradient based pose optimization possible. This shape representation, together with a full perspective projection model, enables more accurate hand modeling than a related baseline method from literature. Our method achieves better accuracy than previous methods and runs at 25 fps. We show these improvements both qualitatively and quantitatively on publicly available datasets.
Tasks
Published	2016-02-11
URL	http://arxiv.org/abs/1602.03860v1
PDF	http://arxiv.org/pdf/1602.03860v1.pdf
PWC	https://paperswithcode.com/paper/real-time-hand-tracking-using-a-sum-of
Repo
Framework

Cellular Automata Segmentation of the Boundary between the Compacta of Vertebral Bodies and Surrounding Structures


Title	Cellular Automata Segmentation of the Boundary between the Compacta of Vertebral Bodies and Surrounding Structures
Authors	Jan Egger, Christopher Nimsky
Abstract	Due to the aging population, spinal diseases get more and more common nowadays; e.g., lifetime risk of osteoporotic fracture is 40% for white women and 13% for white men in the United States. Thus the numbers of surgical spinal procedures are also increasing with the aging population and precise diagnosis plays a vital role in reducing complication and recurrence of symptoms. Spinal imaging of vertebral column is a tedious process subjected to interpretation errors. In this contribution, we aim to reduce time and error for vertebral interpretation by applying and studying the GrowCut-algorithm for boundary segmentation between vertebral body compacta and surrounding structures. GrowCut is a competitive region growing algorithm using cellular automata. For our study, vertebral T2-weighted Magnetic Resonance Imaging (MRI) scans were first manually outlined by neurosurgeons. Then, the vertebral bodies were segmented in the medical images by a GrowCut-trained physician using the semi-automated GrowCut-algorithm. Afterwards, results of both segmentation processes were compared using the Dice Similarity Coefficient (DSC) and the Hausdorff Distance (HD) which yielded to a DSC of 82.99+/-5.03% and a HD of 18.91+/-7.2 voxel, respectively. In addition, the times have been measured during the manual and the GrowCut segmentations, showing that a GrowCut-segmentation - with an average time of less than six minutes (5.77+/-0.73) - is significantly shorter than a pure manual outlining.
Tasks
Published	2016-03-03
URL	http://arxiv.org/abs/1603.00960v1
PDF	http://arxiv.org/pdf/1603.00960v1.pdf
PWC	https://paperswithcode.com/paper/cellular-automata-segmentation-of-the
Repo
Framework

Random Forest Based Approach for Concept Drift Handling


Title	Random Forest Based Approach for Concept Drift Handling
Authors	A. Zhukov, D. Sidorov, A. Foley
Abstract	Concept drift has potential in smart grid analysis because the socio-economic behaviour of consumers is not governed by the laws of physics. Likewise there are also applications in wind power forecasting. In this paper we present decision tree ensemble classification method based on the Random Forest algorithm for concept drift. The weighted majority voting ensemble aggregation rule is employed based on the ideas of Accuracy Weighted Ensemble (AWE) method. Base learner weight in our case is computed for each sample evaluation using base learners accuracy and intrinsic proximity measure of Random Forest. Our algorithm exploits both temporal weighting of samples and ensemble pruning as a forgetting strategy. We present results of empirical comparison of our method with original random forest with incorporated “replace-the-looser” forgetting andother state-of-the-art concept-drfit classifiers like AWE2.
Tasks
Published	2016-02-14
URL	http://arxiv.org/abs/1602.04435v1
PDF	http://arxiv.org/pdf/1602.04435v1.pdf
PWC	https://paperswithcode.com/paper/random-forest-based-approach-for-concept
Repo
Framework

Orientation-boosted Voxel Nets for 3D Object Recognition


Title	Orientation-boosted Voxel Nets for 3D Object Recognition
Authors	Nima Sedaghat, Mohammadreza Zolfaghari, Ehsan Amiri, Thomas Brox
Abstract	Recent work has shown good recognition results in 3D object recognition using 3D convolutional networks. In this paper, we show that the object orientation plays an important role in 3D recognition. More specifically, we argue that objects induce different features in the network under rotation. Thus, we approach the category-level classification task as a multi-task problem, in which the network is trained to predict the pose of the object in addition to the class label as a parallel task. We show that this yields significant improvements in the classification results. We test our suggested architecture on several datasets representing various 3D data sources: LiDAR data, CAD models, and RGB-D images. We report state-of-the-art results on classification as well as significant improvements in precision and speed over the baseline on 3D detection.
Tasks	3D Object Recognition, Object Recognition
Published	2016-04-12
URL	http://arxiv.org/abs/1604.03351v2
PDF	http://arxiv.org/pdf/1604.03351v2.pdf
PWC	https://paperswithcode.com/paper/orientation-boosted-voxel-nets-for-3d-object
Repo
Framework