Many applications in Multilingual and Multimodal Information Access involve searching large databases of high dimensional data objects with multiple (conditionally independent) views. In this work we consider the problem of learning hash functions for similarity search across the views for such applications. We propose a principled method for learning a hash function for each view given a set of multiview training data objects. The hash functions map similar objects to similar codes across the views thus enabling cross-view similarity search. We present results from an extensive empirical study of the proposed approach which demonstrate its effectiveness on Japanese language People Search and Multilingual People Search problems.

The Slow Feature Analysis (SFA) unsupervised learning framework extracts features representing the underlying causes of the changes within a temporally coherent high-dimensional raw sensory input signal. We develop the first online version of SFA, via a combination of incremental Principal Components Analysis and Minor Components Analysis. Unlike standard batch-based SFA, online SFA adapts along with non-stationary environments, which makes it a generally useful unsupervised preprocessor for autonomous learning agents. We compare online SFA to batch SFA in several experiments and show that it indeed learns without a teacher to encode the input stream by informative slow features representing meaningful abstract environmental properties. We extend online SFA to deep networks in hierarchical fashion, and use them to successfully extract abstract object position information from high-dimensional video.

This paper shows how to learn general, Finite State Machine representations of activities that function as recognizers of previously unseen instances of activities. The central problem is to tell which differences between instances of activities are unimportant and may be safely ignored for the purpose of learning generalized representations of activities. We develop a novel way to find the "essential parts" of activities by a greedy kind of multiple sequence alignment, and a method to transform the resulting alignments into Finite State Machine that will accept novel instances of activities with high accuracy.

We investigate the problem of mining numerical data with Formal Concept Analysis. The usual way is to use a scaling procedure —transforming numerical attributes into binary ones — leading either to a loss of information or of efficiency, in particular w.r.t. the volume of extracted patterns. By contrast, we propose to directly work on numerical data in a more precise and efficient way. For that, the notions of closed patterns, generators and equivalent classes are revisited in the numerical context. Moreover, two original algorithms are proposed and tested in an evaluation involving real-world data, showing the quality of the present approach.

The mixture of multivariate Bernoulli distributions (MMB) is a statistical model for high-dimensional binary data in widespread use. Recently, the MMB has been used to model the sequence of packet receptions and losses of wireless links in sensor networks. Given an MMB trained on long data traces recorded from links of a deployed network, one can then use samples from the MMB to test different routing algorithms for as long as desired. However, learning an accurate model for a new link requires collecting from it long traces over periods of hours, a costly process in practice (e.g. limited battery life). We propose an algorithm that can adapt a preexisting MMB trained with extensive data to a new link from which very limited data is available. Our approach constrains the new MMB's parameters through a nonlinear transformation of the existing MMB's parameters. The transformation has a small number of parameters that are estimated using a generalized EM algorithm with an inner loop of BFGS iterations. We demonstrate the efficacy of the approach using the MNIST dataset of handwritten digits, and wireless link data from a sensor network. We show we can learn accurate models from data traces of about 1 minute, about 10 times shorter than needed if training an MMB from scratch.

In this paper, we propose a novel approach for learning regression rules by transforming the regression problem into a classification problem. Unlike previous approaches to regression by classification, in our approach the discretization of the class variable is tightly integrated into the rule learning algorithm. The key idea is to dynamically define a region around the target value predicted by the rule, and considering all examples within that region as positive and all examples outside that region as negative. In this way, conventional rule learning heuristics may be used for inducing regression rules. Our results show that our heuristic algorithm outperforms approaches that use a static discretization of the target variable, and performs en par with other comparable rule-based approaches, albeit without reaching the performance of statistical approaches.

We conjecture that the distribution of the time-reversed residuals of a causal linear process is closer to a Gaussian than the distribution of the noise used to generate the process in the forward direction. This property is demonstrated for causal AR(1) processes assuming that all the cumulants of the distribution of the noise are defined. Based on this observation, it is possible to design a decision rule for detecting the direction of time series that can be described as linear processes: The correct direction (forward in time) is the one in which the residuals from a linear fit to the time series are less Gaussian. A series of experiments with simulated and real-world data illustrate the superior results of the proposed rule when compared with other state-of-the-art methods based on independence tests.

There are a number of papers that use hill-climbing or k-We introduce a new nearest neighbor search algorithm. NN graphs for nearest neighbor search, but to the best of our The algorithm builds a nearest neighbor knowledge, using hill-climbing on k-NN graphs is a new idea.

Most of the sequential patterns extraction methods proposed so far deal with patterns composed of events linked by temporal relationships based on simple precedence between instants. In many real situations, some quantitative information about event duration or inter-event delay is necessary to discriminate phenomena. We propose the algorithm QTIPrefixSpan for extracting temporal patterns composed of events to which temporal intervals describing their position in time and their duration are associated. It extends algorithm PrefixSpan with a multi-dimensional interval clustering step for extracting the representative temporal intervals associated to events in patterns. Experiments on simulated data show that our algorithm is efficient for extracting precise patterns even in noisy contexts and that it improves the performance of a former algorithm which used a clustering method based on the EM algorithm.

In this paper, we tackle the challenges of multi-label classification by developing a general conditional dependency network model. The proposed model is a cyclic directed graphical model, which provides an intuitive representation for the dependencies among multiple label variables, and a well integrated framework for efficient model training using binary classifiers and label predictions using Gibbs sampling inference. Our experiments show the proposed conditional model can effectively exploit the label dependency to improve multi-label classification performance.