We propose a structured prediction algorithm for object localization based on Support Vector Machines (SVMs) using privileged information. Privileged information provides useful high-level knowledge for image understanding and facilitates learning a reliable model even with a small number of training examples. In our setting, we assume that such information is available only at training time since it may be difficult to obtain from visual data accurately without human supervision. Our goal is to improve performance by incorporating privileged information into ordinary learning framework and adjusting model parameters for better generalization. We tackle object localization problem based on a novel structural SVM using privileged information, where an alternating loss-augmented inference procedure is employed to handle the term in the objective function corresponding to privileged information. We apply the proposed algorithm to the Caltech-UCSD Birds 200-2011 dataset, and obtain encouraging results suggesting further investigation into the benefit of privileged information in structured prediction.

To cope with the high level of ambiguity faced in domains such as Computer Vision or Natural Language processing, robust prediction methods often search for a diverse set of high-quality candidate solutions or proposals. In structured prediction problems, this becomes a daunting task, as the solution space (image labelings, sentence parses, etc.) is exponentially large. We study greedy algorithms for finding a diverse subset of solutions in structured-output spaces by drawing new connections between submodular functions over combinatorial item sets and High-Order Potentials (HOPs) studied for graphical models. Specifically, we show via examples that when marginal gains of submodular diversity functions allow structured representations, this enables efficient (sub-linear time) approximate maximization by reducing the greedy augmentation step to inference in a factor graph with appropriately constructed HOPs. We discuss benefits, tradeoffs, and show that our constructions lead to significantly better proposals.

A well-recognized limitation of kernel learning is the requirement to handle a kernel matrix, whose size is quadratic in the number of training examples. Many methods have been proposed to reduce this computational cost, mostly by using a subset of the kernel matrix entries, or some form of low-rank matrix approximation, or a random projection method. In this paper, we study lower bounds on the error attainable by such methods as a function of the number of entries observed in the kernel matrix or the rank of an approximate kernel matrix. We show that there are kernel learning problems where no such method will lead to non-trivial computational savings. Our results also quantify how the problem difficulty depends on parameters such as the nature of the loss function, the regularization parameter, the norm of the desired predictor, and the kernel matrix rank. Our results also suggest cases where more efficient kernel learning might be possible.

Programming a robot to perform tasks requires training that is beyond the skill level of most individuals. To address this issue, we focus on developing a method that identifies keywords used to convey task knowledge among people and a framework that uses these keywords as conditions for knowledge acquisition by the robot learner. The methodology includes generalizing task modeling and providing a robot learner the ability to learn and improve its skills through accumulated experience gained from interaction with humans. More specifically, the aim of this research addresses the issues of knowledge encoding, acquisition, and retrieval through interactive instance-based learning (IIBL). In interaction studies, the benefit of using such a robot learner is in promoting social behaviors that results from the participant taking on an active role as teacher. Our recent experiment with 33 participants, including 19 typically developing children, and a pilot study with two children with autism spectrum disorder showed that IIBL provides a framework for designing an effective robot learner, and that the robot learner successfully increases the amount of social interactions initiated by the participants.

Dropout training, originally designed for deep neural networks, has been successful on high-dimensional single-layer natural language tasks. This paper proposes a theoretical explanation for this phenomenon: we show that, under a generative Poisson topic model with long documents, dropout training improves the exponent in the generalization bound for empirical risk minimization. Dropout achieves this gain much like a marathon runner who practices at altitude: once a classifier learns to perform reasonably well on training examples that have been artificially corrupted by dropout, it will do very well on the uncorrupted test set. We also show that, under similar conditions, dropout preserves the Bayes decision boundary and should therefore induce minimal bias in high dimensions.

The ever-increasing size of modern data sets combined with the difficulty of obtaining label information has made semi-supervised learning one of the problems of significant practical importance in modern data analysis. We revisit the approach to semi-supervised learning with generative models and develop new models that allow for effective generalisation from small labelled data sets to large unlabelled ones. Generative approaches have thus far been either inflexible, inefficient or non-scalable. We show that deep generative models and approximate Bayesian inference exploiting recent advances in variational methods can be used to provide significant improvements, making generative approaches highly competitive for semi-supervised learning.

In this paper we analyze a budgeted learning setting, in which the learner can only choose and observe a small subset of the attributes of each training example. We develop efficient algorithms for ridge and lasso linear regression, which utilize the geometry of the data by a novel data-dependent sampling scheme. When the learner has prior knowledge on the second moments of the attributes, the optimal sampling probabilities can be calculated precisely, and result in data-dependent improvements factors for the excess risk over the state-of-the-art that may be as large as $O(\sqrt{d})$, where $d$ is the problem's dimension. Moreover, under reasonable assumptions our algorithms can use less attributes than full-information algorithms, which is the main concern in budgeted learning settings. To the best of our knowledge, these are the first algorithms able to do so in our setting. Where no such prior knowledge is available, we develop a simple estimation technique that given a sufficient amount of training examples, achieves similar improvements. We complement our theoretical analysis with experiments on several data sets which support our claims.

We present a novel approach to learn binary classifiers when only positive and unlabeled instances are available (PU learning). This problem is routinely cast as a supervised task with label noise in the negative set. We use an ensemble of SVM models trained on bootstrap resamples of the training data for increased robustness against label noise. The approach can be considered in a bagging framework which provides an intuitive explanation for its mechanics in a semi-supervised setting. We compared our method to state-of-the-art approaches in simulations using multiple public benchmark data sets. The included benchmark comprises three settings with increasing label noise: (i) fully supervised, (ii) PU learning and (iii) PU learning with false positives. Our approach shows a marginal improvement over existing methods in the second setting and a significant improvement in the third. Frank De Smet is a member of the medical management department of the National Alliance of Christian Mutualities. Accepted at Neurocomputing: SI on Advances in Learning with Label Noise 20/10/2014 1. Introduction Training binary classifiers on positive and unlabeled data is referred to as PU learning [31]. The absence of known negative training instances warrants appropriate learning methods. Inaccurate label information can be more problematic than attribute noise [45]. Specialised PU learning approaches are recommended when (i) negative labels cannot be acquired, (ii) the training data contains a large amount of false negatives or (iii) the positive set has many outliers. Practical applications of PU learning typically feature large, imbalanced training sets with a small amount of labeled (positive) and a large amount of unlabeled training instances. The PU learning problem arises in various settings, including web page classification [44], intrusion detection [26] and bioinformatics tasks such as variant prioritization [42], gene prioritization [1, 35] and virtual screening of drug compounds [41]. Though these applications share a common underlying learning problem, the final evaluation criteria may be fundamentally different.

Objective. The paper investigates the presence of autism using the functional brain connectivity measures derived from electro-encephalogram (EEG) of children during face perception tasks. Approach. Phase synchronized patterns from 128-channel EEG signals are obtained for typical children and children with autism spectrum disorder (ASD). The phase synchronized states or synchrostates temporally switch amongst themselves as an underlying process for the completion of a particular cognitive task. We used 12 subjects in each group (ASD and typical) for analyzing their EEG while processing fearful, happy and neutral faces. The minimal and maximally occurring synchrostates for each subject are chosen for extraction of brain connectivity features, which are used for classification between these two groups of subjects. Among different supervised learning techniques, we here explored the discriminant analysis and support vector machine both with polynomial kernels for the classification task. Main results. The leave one out cross-validation of the classification algorithm gives 94.7% accuracy as the best performance with corresponding sensitivity and specificity values as 85.7% and 100% respectively. Significance. The proposed method gives high classification accuracies and outperforms other contemporary research results. The effectiveness of the proposed method for classification of autistic and typical children suggests the possibility of using it on a larger population to validate it for clinical practice.

Mach Learn manuscript No. (will be inserted by the editor) Abstract In this paper, we consider a supervised learning setting where side knowledge is provided about the labels of unlabeled examples. The side knowledge has the effect of reducing the hypothesis space, leading to tighter generalization bounds, and thus possibly better generalization. We consider several types of side knowledge, the first leading to linear and polygonal constraints on the hypothesis space, the second leading to quadratic constraints, and the last leading to conic constraints. We show how different types of domain knowledge can lead directly to these kinds of side knowledge. We prove bounds on complexity measures of the hypothesis space for quadratic and conic side knowledge, and show that these bounds are tight in a specific sense for the quadratic case. Keywords statistical learning theory · generalization bounds · Rademacher complexity · covering numbers, constrained linear function classes · side knowledge 1 Introduction Surely, for many applications the amount of domain knowledge we could potentially use within our learning processes is vastly larger than the amount of domain knowledge we actually use. One reason for this is that domain knowledge may be nontrivial to incorporate into algorithms or analysis. For example, researchers in NLP (Natural Language Processing) have long figured out various exotic domain specific knowledge that one can use while performing a learning task [Chang et al., 2008a,b]. The present work aims to provide theoretical guarantees for a large class of problems with a general type of domain knowledge that goes beyond sparsity and smoothness. To define this large class of problems, we will keep the usual supervised learning assumption that the training examples are drawn i.i.d.