When doing learning from demonstration, it is often the case that the demonstrator provides corrective examples to fix errant behavior by the agent or robot. We present a set of algorithms which use this corrective data to identify and remove noisy examples in datasets which caused errant classifications, and ultimately errant behavior. The objective is to actually modify the source datasets rather than solely rely on the noise-insensitivity of the classification algorithm. This is particularly useful in the sparse datasets often found in learning from demonstration experiments. Our approach tries to distinguish between noisy misclassification and mere undersampling of the learning space. If errors are a result of misclassification, we potentially remove the responsible points and update the classifier. We demonstrate our method on UCI Machine Learning datasets at different levels of sparsity and noise, using decision trees, K-Nearest-Neighbor, and support vector machines.

We propose a new approach, multi-view Laplacian support vector machines (SVMs), for semi-supervised learning under the multi-view scenario. It integrates manifold regularization and multi-view regularization into the usual formulation of SVMs and is a natural extension of SVMs from supervised learning to multi-view semi-supervised learning. The function optimization problem in a reproducing kernel Hilbert space is converted to an optimization in a finite-dimensional Euclidean space. After providing a theoretical bound for the generalization performance of the proposed method, we further give a formulation of the empirical Rademacher complexity which affects the bound significantly. From this bound and the empirical Rademacher complexity, we can gain insights into the roles played by different regularization terms to the generalization performance. Experimental results on synthetic and real-world data sets are presented, which validate the effectiveness of the proposed multi-view Laplacian SVMs approach.

The crucial importance of metrics in machine learning algorithms has led to an increasing interest in optimizing distance and similarity functions, an area of research known as metric learning. When data consist of feature vectors, a large body of work has focused on learning a Mahalanobis distance. Less work has been devoted to metric learning from structured objects (such as strings or trees), most of it focusing on optimizing a notion of edit distance. We identify two important limitations of current metric learning approaches. First, they allow to improve the performance of local algorithms such as k-nearest neighbors, but metric learning for global algorithms (such as linear classifiers) has not been studied so far. Second, the question of the generalization ability of metric learning methods has been largely ignored. In this thesis, we propose theoretical and algorithmic contributions that address these limitations. Our first contribution is the derivation of a new kernel function built from learned edit probabilities. Our second contribution is a novel framework for learning string and tree edit similarities inspired by the recent theory of (e,g,t)-good similarity functions. Using uniform stability arguments, we establish theoretical guarantees for the learned similarity that give a bound on the generalization error of a linear classifier built from that similarity. In our third contribution, we extend these ideas to metric learning from feature vectors by proposing a bilinear similarity learning method that efficiently optimizes the (e,g,t)-goodness. Generalization guarantees are derived for our approach, highlighting that our method minimizes a tighter bound on the generalization error of the classifier. Our last contribution is a framework for establishing generalization bounds for a large class of existing metric learning algorithms based on a notion of algorithmic robustness.

We introduce a conceptually novel structured prediction model, GPstruct, which is kernelized, non-parametric and Bayesian, by design. We motivate the model with respect to existing approaches, among others, conditional random fields (CRFs), maximum margin Markov networks (M3N), and structured support vector machines (SVMstruct), which embody only a subset of its properties. We present an inference procedure based on Markov Chain Monte Carlo. The framework can be instantiated for a wide range of structured objects such as linear chains, trees, grids, and other general graphs. As a proof of concept, the model is benchmarked on several natural language processing tasks and a video gesture segmentation task involving a linear chain structure. We show prediction accuracies for GPstruct which are comparable to or exceeding those of CRFs and SVMstruct.

We build turn-taking models for autonomous characters in language-based interactions with small groups of children. Two models explore the use of support vector machines given the same multimodal features, but different methods for collecting turn-taking labels.

Detection of cognitive impairment, especially at the early stages, is critical. Such detection has traditionally been performed manually by one or more clinicians based on reports and test results. Machine learning algorithms offer an alternative method of detection that may provide an automated process and valuable insights into diagnosis and classification. In this paper, we explore the use of neuropsychological and demographic data to predict Clinical Dementia Rating (CDR) scores (no dementia, very mild dementia, dementia) and clinical diagnoses (cognitively healthy, mild cognitive impairment, dementia) through the implementation of four machine learning algorithms, naïve Bayes (NB), C4.5 decision tree (DT), back-propagation neural network (NN), and support vector machine (SVM). Additionally, a feature selection method for reducing the number of neuropsychological and demographic data needed to make an accurate diagnosis was investigated. The NB classifier provided the best accuracies, while the SVM classifier proved to offer some of the lowest accuracies. We also illustrate that with the use of feature selection, accuracies can be improved. The experiments reported in this paper indicate that artificial intelligence techniques can be used to automate aspects of clinical diagnosis of individuals with cognitive impairment.

While recent Machine Learning (ML) based techniques for activity recognition show great promise, there remain a number of questions with respect to the relative merits of these techniques. To provide a better understanding of the relative strengths of contemporary Activity Recognition methods, in this paper we present a comparative analysis of Hidden Markov Model, Bayesian, and Support Vector Machine based human activity recognition models. The study builds on both pre-existing and newly annotated data which includes interleaved activities. Results demonstrate that while Support Vector Machine based techniques perform well for all data sets considered, simple representations of sensor histories regularly outperform more complex count based models.

One-against-all and one-against-one are two popular methodologies for reducing multiclass classification problems into a set of binary classifications. In this paper, we are interested in the performance of both one-against-all and one-against-one for classification algorithms, such as decision tree, naïve bayes, support vector machine, and logistic regression. Since both one-against-all and one-against-one work like creating a classification committee, they are expected to improve the performance of classification algorithms. However, our experimental results surprisingly show that one-against-all worsens the performance of the algorithms on most datasets. One-against-one helps, but performs worse than the same iterations of bagging these algorithms. Thus, we conclude that both one-against-all and one-against-one should not be used for the algorithms that can perform multiclass classifications directly. Bagging is better approach for improving their performance.

For the natural and social interaction it is necessary to understand human behavior. Personality is one of the fundamental aspects, by which we can understand behavioral dispositions. It is evident that there is a strong correlation between users’ personality and the way they behave on online social network (e.g., Facebook). This paper presents automatic recognition of Big-5 personality traits on social network (Facebook) using users’ status text. For the automatic recognition we studied different classification methods such as SMO (Sequential Minimal Optimization for Support Vector Machine), Bayesian Logistic Regression (BLR) and Multinomial Naïve Bayes (MNB) sparse modeling. Performance of the systems had been measured using macro-averaged precision, recall and F1; weighted average accuracy (WA) and un-weighted average accuracy (UA). Our comparative study shows that MNB performs better than BLR and SMO for personality traits recognition on the social network data.

Machine Learning (ML) algorithms are used to train computers to perform a variety of complex tasks and improve with experience. Computers learn how to recognize patterns, make unintended decisions, or react to a dynamic environment. Certain trained machines may be more effective than others because they are based on more suitable ML algorithms or because they were trained through superior training sets. Although ML algorithms are known and publicly released, training sets may not be reasonably ascertainable and, indeed, may be guarded as trade secrets. While much research has been performed about the privacy of the elements of training sets, in this paper we focus our attention on ML classifiers and on the statistical information that can be unconsciously or maliciously revealed from them. We show that it is possible to infer unexpected but useful information from ML classifiers. In particular, we build a novel meta-classifier and train it to hack other classifiers, obtaining meaningful information about their training sets. This kind of information leakage can be exploited, for example, by a vendor to build more effective classifiers or to simply acquire trade secrets from a competitor's apparatus, potentially violating its intellectual property rights.