Creating descriptors for trajectories has many applications in robotics/human motion analysis and video copy detection. Here, we propose a novel descriptor for 2D trajectories: Histogram of Oriented Displacements (HOD). Each displacement in the trajectory votes with its length in a histogram of orientation angles. 3D trajectories are described by the HOD of their three projections. We use HOD to describe the 3D trajectories of body joints to recognize human actions, which is a challenging machine vision task, with applications in human-robot/machine interaction, interactive entertainment, multimedia information retrieval, and surveillance. The descriptor is fixed-length, scale-invariant and speed-invariant. Experiments on MSR-Action3D and HDM05 datasets show that the descriptor outperforms the state-of-the-art when using off-the-shelf classification tools.

Many information gathering problems require determining the set of points, for which an unknown function takes value above or below some given threshold level. We formalize this task as a classification problem with sequential measurements, where the unknown function is modeled as a sample from a Gaussian process (GP). We propose LSE, an algorithm that guides both sampling and classification based on GP-derived confidence bounds, and provide theoretical guarantees about its sample complexity. Furthermore, we extend LSE and its theory to two more natural settings: (1) where the threshold level is implicitly defined as a percentage of the (unknown) maximum of the target function and (2) where samples are selected in batches. We evaluate the effectiveness of our proposed methods on two problems of practical interest, namely autonomous monitoring of algal populations in a lake environment and geolocating network latency.

During the past decade, finite mixture modeling has become a well-established technique in data analysis and clustering. This paper focus on developing a variational inference framework to learn finite Beta-Liouville mixture models that have been proposed recently as an efficient way for proportional data clustering. In contrast to the conventional expectation maximization (EM) algorithm, commonly used for learning finite mixture models, the proposed algorithm has the advantages that it is more efficient from a computational point of view and by preventing over- and under-fitting problems. Moreover, the complexity of the mixture model (i.e. the number of components) can be determined automatically and simultaneously with the parameters estimation in a closed form as part of the Bayesian inference procedure. The merits of the proposed approach are shown using both artificial data sets and two interesting and challenging real applications namely dynamic textures clustering and facial expression recognition.

In many problem settings, for example on graph domains, online learning algorithms on streams of data need to respect strict time constraints dictated by the throughput on which the data arrive. When only a limited amount of memory (budget) is available, a learning algorithm will eventually need to discard some of the information used to represent the current solution, thus negatively affecting its classification performance. More importantly, the overhead due to budget management may significantly increase the computational burden of the learning algorithm. In this paper we present a novel approach inspired by the Passive Aggressive and the Lossy Counting algorithms. Our algorithm uses a fast procedure for deleting the less influential features. Moreover, it is able to estimate the weighted frequency of each feature and use it for prediction.

Most of the algorithms for inverse reinforcement learning (IRL) assume that the reward function is a linear function of the pre-defined state and action features. However, it is often difficult to manually specify the set of features that can make the true reward function representable as a linear function. We propose a Bayesian nonparametric approach to identifying useful composite features for learning the reward function. The composite features are assumed to be the logical conjunctions of the predefined atomic features so that we can represent the reward function as a linear function of the composite features. We empirically show that our approach is able to learn composite features that capture important aspects of the reward function on synthetic domains, and predict taxi drivers’ behaviour with high accuracy on a real GPS trace dataset.

A serious and ubiquitous issue in machine learning is the lack ofsufficient training data in a domain of interest. Domain adaptation isan effective approach to dealing with this problem by transferringinformation or models learned from related, albeit distinct, domains tothe target domain. We develop a novel domain adaptation method fortext document classification under the framework of Non-negativeMatrix Factorization. Two key ideas of our method are to construct alatent topic space where a topic is decomposed into common wordsshared by all domains and words specific to individual domains, andthen to establish associations between words in different domainsthrough the common words as a bridge for knowledge transfer. Thecorrespondence between cross-domain topics leads to more coherentdistributions of source and target domains in the new representationwhile preserving the predictive power. Our new method outperformedseveral state-of-the-art domain adaptation methods on severalbenchmark datasets.

Relational topic models have shown promise on analyzing document network structures and discovering latent topic representations. This paper presents three extensions: 1) unlike the common link likelihood with a diagonal weight matrix that allows the-same-topic interactions only, we generalize it to use a full weight matrix that captures all pairwise topic interactions and is applicable to asymmetric networks; 2) instead of doing standard Bayesian inference, we perform regularized Bayesian inference with a regularization parameter to deal with the imbalanced link structure issue in common real networks; and 3) instead of doing variational approximation with strict mean-field assumptions, we present a collapsed Gibbs sampling algorithm for the generalized relational topic models without making restricting assumptions. Experimental results demonstrate the significance of these extensions on improving the prediction performance, and the time efficiency can be dramatically improved with a simple fast approximation method.

In this paper, we propose a general dimensionality reduction method for data generated from a very broad family of distributions and nonlinear functions based on the generalized linear model, called Generalized Linear Principal Component Analysis (GLPCA). Data of different domains often have very different structures. These data can be modeled by different distributions and reconstruction functions. For example, real valued data can be modeled by the Gaussian distribution with a linear reconstruction function, whereas binary valued data may be more appropriately modeled by the Bernoulli distribution with a logit or probit function. Based on general linear models, we propose a unified framework for extracting features from data of different domains. A general optimization algorithm based on natural gradient ascent on distribution manifold is proposed for obtaining the maximum likelihood solutions. We also present some specific algorithms derived from this framework to deal with specific data modeling problems such as document modeling. Experimental results of these algorithms on several data sets are shown for the validation of GLPCA.

Tensors are increasingly common in several areas such as data mining, computer graphics, and computer vision. Tensor clustering is a fundamental tool for data analysis and pattern discovery. However, there usually exist outlying data points in realworld datasets, which will reduce the performance of clustering. This motivates us to develop a tensor clustering algorithm that is robust to the outliers. In this paper, we propose an algorithm of Robust Tensor Clustering (RTC). The RTC firstly finds a lowerrank approximation of the original tensor data using a L1 norm optimization function. Because the L1 norm doesn’t exaggerate the effect of outliers compared with L2 norm, the minimization of the L1 norm approximation function makes RTC robust to outliers. Then we compute the HOSVD decomposition of this approximate tensor to obtain the final clustering results. Different from the traditional algorithm solving the approximation function with a greedy strategy, we utilize a non-greedy strategy to obtain a better solution. Experiments demonstrate that RTC has better performance than the state-ofthe-art algorithms and is more robust to outliers.

Pose variation is one of the challenging factors for face recognition. In this paper, we propose a novel cross-pose face recognition method named as Regularized Latent Least Square Regression (RLLSR). The basic assumption is that the images captured under different poses of one person can be viewed as pose-specific transforms of a single ideal object. We treat the observed images as regressor, the ideal object as response, and then formulate this assumption in the least square regression framework, so as to learn the multiple pose-specific transforms. Specifically, we incorporate some prior knowledge as two regularization terms into the least square approach: 1) the smoothness regularization, as the transforms for nearby poses should not differ too much; 2) the local consistency constraint, as the distribution of the latent ideal objects should preserve the geometric structure of the observed image space. We develop an alternating algorithm to simultaneously solve for the ideal objects of the training individuals and a set of pose-specific transforms. The experimental results on the Multi-PIE dataset demonstrate the effectiveness of the proposed method and superiority over the previous methods.