Accelerometer measurements are the prime type of sensor information most think of when seeking to measure physical activity. On the market, there are many fitness measuring devices which aim to track calories burned and steps counted through the use of accelerometers. These measurements, though good enough for the average consumer, are noisy and unreliable in terms of the precision of measurement needed in a scientific setting. The contribution of this paper is an innovative and highly accurate regression method which uses an intermediary two-stage classification step to better direct the regression of energy expenditure values from accelerometer counts. We show that through an additional unsupervised layer of intermediate feature construction, we can leverage latent patterns within accelerometer counts to provide better grounds for activity classification than expert-constructed timeseries features. For this, our approach utilizes a mathematical model originating in natural language processing, the bag-of-words model, that has in the past years been appearing in diverse disciplines outside of the natural language processing field such as image processing. Further emphasizing the natural language connection to stochastics, we use a gaussian mixture model to learn the dictionary upon which the bag-of-words model is built. Moreover, we show that with the addition of these features, we're able to improve regression root mean-squared error of energy expenditure by approximately 1.4 units over existing state-of-the-art methods.

Bayesian network structure learning is often performed in a Bayesian setting, evaluating candidate structures using their posterior probabilities for a given data set. Score-based algorithms then use those posterior probabilities as an objective function and return the maximum a posteriori network as the learned model. For discrete Bayesian networks, the canonical choice for a posterior score is the Bayesian Dirichlet equivalent uniform (BDeu) marginal likelihood with a uniform (U) graph prior, which assumes a uniform prior both on the network structures and on the parameters of the networks. In this paper, we investigate the problems arising from these assumptions, focusing on those caused by small sample sizes and sparse data. We then propose an alternative posterior score: the Bayesian Dirichlet sparse (BDs) marginal likelihood with a marginal uniform (MU) graph prior. Like U BDeu, MU BDs does not require any prior information on the probabilistic structure of the data and can be used as a replacement noninformative score. We study its theoretical properties and we evaluate its performance in an extensive simulation study, showing that MU BDs is both more accurate than U BDeu in learning the structure of the network and competitive in predicting power, while not being computationally more complex to estimate.

Many computer vision tasks involve processing large amounts of data contaminated by outliers, which need to be detected and rejected. While outlier detection methods based on robust statistics have existed for decades, only recently have methods based on sparse and low-rank representation been developed along with guarantees of correct outlier detection when the inliers lie in one or more low-dimensional subspaces. This paper proposes a new outlier detection method that combines tools from sparse representation with random walks on a graph. By exploiting the property that data points can be expressed as sparse linear combinations of each other, we obtain an asymmetric affinity matrix among data points, which we use to construct a weighted directed graph. By defining a suitable Markov Chain from this graph, we establish a connection between inliers/outliers and essential/inessential states of the Markov chain, which allows us to detect outliers by using random walks. We provide a theoretical analysis that justifies the correctness of our method under geometric and connectivity assumptions. Experimental results on image databases demonstrate its superiority with respect to state-of-the-art sparse and low-rank outlier detection methods.

The task of determining item similarity is a crucial one in a recommender system. This constitutes the base upon which the recommender system will work to determine which items are more likely to be enjoyed by a user, resulting in more user engagement. In this paper we tackle the problem of determining song similarity based solely on song metadata (such as the performer, and song title) and on tags contributed by users. We evaluate our approach under a series of different machine learning algorithms. We conclude that tf-idf achieves better results than Word2Vec to model the dataset to feature vectors. We also conclude that k-NN models have better performance than SVMs and Linear Regression for this problem.

Feature extraction is a very crucial task in image and pixel (voxel) classification and regression in biomedical image modeling. In this work we present a machine learning based feature extraction scheme based on inception models for pixel classification tasks. We extract features under multi-scale and multi-layer schemes through convolutional operators. Layers of Fully Convolutional Network are later stacked on this feature extraction layers and trained end-to-end for the purpose of classification. We test our model on the DRIVE and STARE public data sets for the purpose of segmentation and centerline detection and it out performs most existing hand crafted or deterministic feature schemes found in literature. We achieve an average maximum Dice of 0.85 on the DRIVE data set which out performs the scores from the second human annotator of this data set. We also achieve an average maximum Dice of 0.85 and kappa of 0.84 on the STARE data set. Though these datasets are mainly 2-D we also propose ways of extending this feature extraction scheme to handle 3-D datasets.

Work on Automatic Music Transcription (AMT) dates back more than 30 years, and has known numerous applications in the fields of music information retrieval, interactive computer systems, and automated musicological analysis (Klapuri, 2004). Due to the difficulty in producing all the information required for a complete musical score, AMT is commonly defined as the computer-assisted process of analyzing an acoustic musical signal so as to write down the musical parameters of the sounds that occur in it, which are basically the pitch, onset time, and duration of each sound to be played. Despite a large enthusiasm for AMT challenges, and several audio-to-MIDI converters available commercially, perfect polyphonic AMT systems are out of reach of today's technology (Klapuri, 2004; Benetos et al., 2013b). To overcome these limitations, a practical engineering solution was to use computational techniques from statistics and digital signal processing, allowing more complex modeling of the musical signal. In this paper, we investigate the use of different Hidden Markov Models (HMMs) in AMT, and evaluate their impacts on transcription performance. HMMs are a ubiquitous tool to model time series data, and have been widely used in various tasks of Music Information Retrieval, especially in music structure analysis by characterizing repetitive patterns (Logan and Chu, 2000) or performing harmonic analysis (Raphael and Stoddard, 2003), chord estimation (Lee and Slaney, 2008) and musicological modeling of note transitions (Ryynanen and Klapuri, 2008). For what concerns the task of AMT, the sequential structure that may be inferred from musical signals can be usefully integrated to systems with HMMs.

Bayesian optimization (BO) has emerged during the last few years as an effective approach to optimizing black-box functions where direct queries of the objective are expensive. In this paper we consider the case where direct access to the function is not possible, but information about user preferences is. Such scenarios arise in problems where human preferences are modeled, such as A/B tests or recommender systems. We present a new framework for this scenario that we call Preferential Bayesian Optimization (PBO) which allows us to find the optimum of a latent function that can only be queried through pairwise comparisons, the so-called duels. PBO extends the applicability of standard BO ideas and generalizes previous discrete dueling approaches by modeling the probability of the winner of each duel by means of a Gaussian process model with a Bernoulli likelihood. The latent preference function is used to define a family of acquisition functions that extend usual policies used in BO. We illustrate the benefits of PBO in a variety of experiments, showing that PBO needs drastically fewer comparisons for finding the optimum. According to our experiments, the way of modeling correlations in PBO is key in obtaining this advantage.

With the recent development in mobile computing devices and as the ubiquitous deployment of access points(APs) of Wireless Local Area Networks(WLANs), WLAN based indoor localization systems(WILSs) are of mounting concentration and are becoming more and more prevalent for they do not require additional infrastructure. As to the localization methods in WILSs, for the approaches used to localization in satellite based global position systems are difficult to achieve in indoor environments, fingerprint based localization algorithms(FLAs) are predominant in the RSS based schemes. However, the performance of FLAs has close relationship with the number of APs and the number of reference points(RPs) in WILSs, especially as the redundant deployment of APs and RPs in the system. There are two fatal problems, curse of dimensionality (CoD) and asymmetric matching(AM), caused by increasing number of APs and breaking down APs during online stage. In this paper, a semi-supervised RSS dimensionality reduction algorithm is proposed to solve these two dilemmas at the same time and there are numerous analyses about the theoretical realization of the proposed method. Another significant innovation of this paper is jointing the fingerprint based algorithm with CM-SDE algorithm to improve the localization accuracy of indoor localization.

This paper presents sampling-based speech parameter generation using moment-matching networks for Deep Neural Network (DNN)-based speech synthesis. Although people never produce exactly the same speech even if we try to express the same linguistic and para-linguistic information, typical statistical speech synthesis produces completely the same speech, i.e., there is no inter-utterance variation in synthetic speech. To give synthetic speech natural inter-utterance variation, this paper builds DNN acoustic models that make it possible to randomly sample speech parameters. The DNNs are trained so that they make the moments of generated speech parameters close to those of natural speech parameters. Since the variation of speech parameters is compressed into a low-dimensional simple prior noise vector, our algorithm has lower computation cost than direct sampling of speech parameters. As the first step towards generating synthetic speech that has natural inter-utterance variation, this paper investigates whether or not the proposed sampling-based generation deteriorates synthetic speech quality. In evaluation, we compare speech quality of conventional maximum likelihood-based generation and proposed sampling-based generation. The result demonstrates the proposed generation causes no degradation in speech quality.

The Temporal Group LASSO is an example of a multi-task, regularized regression approach for the prediction of response variables that vary over time. The aim of this work is to introduce the reader to the concepts behind the Temporal Group LASSO and its related methods, as well as to the type of potential applications in a healthcare setting that the method has. Weargue thatthemethodis attractivebecause ofitsabilitytoreduce overfitting, select predictors, learn smooth effect patterns over time, and finally, its simplicity.