Topic modeling of textual corpora is an important and challenging problem. In most previous work, the “bag-of-words” assumption is usually made which ignores the ordering of words. This assumption simplifies the computation, but it unrealistically loses the ordering information and the semantic of words in the context. In this paper, we present a Gaussian Mixture Neural Topic Model (GMNTM) which incorporates both the ordering of words and the semantic meaning of sentences into topic modeling. Specifically, we represent each topic as a cluster of multi-dimensional vectors and embed the corpus into a collection of vectors generated by the Gaussian mixture model. Each word is affected not only by its topic, but also by the embedding vector of its surrounding words and the context. The Gaussian mixture components and the topic of documents, sentences and words can be learnt jointly. Extensive experiments show that our model can learn better topics and more accurate word distributions for each topic. Quantitatively, comparing to state-of-the-art topic modeling approaches, GMNTM obtains significantly better performance in terms of perplexity, retrieval accuracy and classification accuracy.

Human gait has been shown to be an efficient biometric measure for person identification at a distance. However, it often needs different gait features to handle various covariate conditions including viewing angles, walking speed, carrying an object and wearing different types of shoes. In order to improve the robustness of gait-based person re-identification on such multi-covariate conditions, a novel Swiss-system based cascade ranking model is proposed in this paper. Since the ranking model is able to learn a subspace where the potential true match is given the highest ranking, we formulate the gait-based person re-identification as a bipartite ranking problem and utilize it as an effective way for multi-feature ensemble learning. Then a Swiss multi-round competition system is developed for the cascade ranking model to optimize its effectiveness and efficiency. Extensive experiments on three indoor and outdoor public datasets demonstrate that our model outperforms several state-of-the-art methods remarkably.

Most word embedding models typically represent each word using a single vector, which makes these models indiscriminative for ubiquitous homonymy and polysemy. In order to enhance discriminativeness, we employ latent topic models to assign topics for each word in the text corpus, and learn topical word embeddings (TWE) based on both words and their topics. In this way, contextual word embeddings can be flexibly obtained to measure contextual word similarity. We can also build document representations, which are more expressive than some widely-used document models such as latent topic models. In the experiments, we evaluate the TWE models on two tasks, contextual word similarity and text classification. The experimental results show that our models outperform typical word embedding models including the multi-prototype version on contextual word similarity, and also exceed latent topic models and other representative document models on text classification.

Multiple kernel learning (MKL) optimally combines the multiple channels of each sample to improve classification performance. However, existing MKL algorithms cannot effectively handle the situation where some channels are missing, which is common in practical applications. This paper proposes an absent MKL (AMKL) algorithm to address this issue. Different from existing approaches where missing channels are firstly imputed and then a standard MKL algorithm is deployed on the imputed data, our algorithm directly classifies each sample with its observed channels. In specific, we define a margin for each sample in its own relevant space, which corresponds to the observed channels of that sample. The proposed AMKL algorithm then maximizes the minimum of all sample-based margins, and this leads to a difficult optimization problem. We show that this problem can be reformulated as a convex one by applying the representer theorem. This makes it readily be solved via existing convex optimization packages. Extensive experiments are conducted on five MKL benchmark data sets to compare the proposed algorithm with existing imputation-based methods. As observed, our algorithm achieves superior performance and the improvement is more significant with the increasing missing ratio.

We consider the problem of learning deep representation when target labels are available. In this paper, we show that there exists intrinsic relationship between target coding and feature representation learning in deep networks. Specifically, we found that distributed binary acode with error correcting capability is more capable of encouraging discriminative features, in comparison tothe 1-of-K coding that is typically used in supervised deep learning. This new finding reveals additional benefit of using error-correcting code for deep model learning,apart from its well-known error correcting property. Extensive experiments are conducted on popular visual benchmark datasets.

While if kernels are not Kernel methods (Schölkopf and Smola 2002; Xu et al. 2009) low rank, Nyström approximations can usually lead to suboptimal have received a lot of attention in recent studies of machine performances. To alleviate the strong assumption in learning. These methods project data into high-dimensional the seeking of the approximation bounds, we take a more or even infinite-dimensional spaces via kernel mapping general assumption that the design matrix K ensuring the restricted functions. Despite the strong generalization ability induced isometric property (Koltchinskii 2011). In particular, by kernel methods, they usually suffer from the high computation the new assumption obeys the restricted eigenvalue condition complexity of calculating the kernel matrix (also (Koltchinskii 2011; Bickel, Ritov, and Tsybakov 2009), called Gram matrix). Although low-rank decomposition which has been shown to be more general than several techniques(e.g., Cholesky Decomposition (Fine and Scheinberg other similar assumptions used in sparsity literature (Candes 2002; Bach and Jordan 2005)), and truncating methods(e.g., and Tao 2007; Donoho, Elad, and Temlyakov 2006; Kernel Tapering (Shen, Xu, and Allebach 2014; Zhang and Huang 2008). Based on the restricted eigenvalue Furrer, Genton, and Nychka 2006)) can accelerate the calculation condition, we have provided error bounds for kernel approximation of the kernel matrix, they still need to compute the and recovery rate in sparse kernel regression.

Active learning is a machine learning technique that trains a classifier after selecting a subset from an unlabeled dataset for labeling and using the selected data for training. Recently, batch mode active learning, which selects a batch of samples to label in parallel, has attracted a lot of attention. Its challenge lies in the choice of criteria used for guiding the search of the optimal batch. In this paper, we propose a novel approach to selecting the optimal batch of queries by minimizing the α-relative Pearson divergence (RPE) between the labeled and the original datasets. This particular divergence is chosen since it can distinguish the optimal batch more easily than other measures especially when available candidates are similar. The proposed objective is a min-max optimization problem, and it is difficult to solve due to the involvement of both minimization and maximization. We find that the objective has an equivalent convex form, and thus a global optimal solution can be obtained. Then the subgradient method can be applied to solve the simplified convex problem. Our empirical studies on UCI datasets demonstrate the effectiveness of the proposed approach compared with the state-of-the-art batch mode active learning methods.

The development and maintenance of large and complex ontologies are often time-consuming and error-prone. Thus, automated ontology learning and evolution have attracted intensive research interest. In data-centric applications where ontologies are designed from the data or automatically learnt from it, when new data instances are added that contradict the ontology, it is often desirable to incrementally revise the ontology according to the added data. In description logics, this problem can be intuitively formulated as the operation of TBox contraction, i.e., rational elimination of certain axioms from the logical consequences of a TBox, and it is w.r.t. an ABox. In this paper we introduce a model-theoretic approach to such a contraction problem by using an alternative semantic characterisation of DL-Lite TBoxes. We show that entailment checking (without necessarily first computing the contraction result) is in coNP, which does not shift the corresponding complexity in propositional logic, and the problem is tractable when the size of the new data is bounded.

Link prediction and multi-label learning on graphs are two important but challenging machine learning problems that have broad applications in diverse fields. Not only are the two problems inherently correlated and often appear concurrently, they are also exacerbated by incomplete data. We develop a novel algorithm to solve these two problems jointly under a unified framework, which helps reduce the impact of graph noise and benefits both tasks individually. We reduce multi-label learning problem into an additional link prediction task and solve both problems with marginalized denoising, which we co-regularize with Laplacian smoothing. This approach combines both learning tasks into a single convex objective function, which we optimize efficiently with iterative closed-form updates. The resulting approach performs significantly better than prior work on several important real-world applications with great consistency.

Learning an appropriate feature representation across source and target domains is one of the most effective solutions to domain adaptation problems. Conventional cross-domain feature learning methods rely on the Reproducing Kernel Hilbert Space (RKHS) induced by a single kernel. Recently, Multiple Kernel Learning (MKL), which bases classifiers on combinations of kernels, has shown improved performance in the tasks without distribution difference between domains. In this paper, we generalize the framework of MKL for cross-domain feature learning and propose a novel Transfer Feature Representation (TFR) algorithm. TFR learns a convex combination of multiple kernels and a linear transformation in a single optimization which integrates the minimization of distribution difference with the preservation of discriminating power across domains. As a result, standard machine learning models trained in the source domain can be reused for the target domain data. After rewritten into a differentiable formulation, TFR can be optimized by a reduced gradient method and reaches the convergence. Experiments in two real-world applications verify the effectiveness of our proposed method.