Margin-based Learning Research Articles

Positive-Unlabeled Learning With Label Distribution Alignment.

Positive-Unlabeled (PU) data arise frequently in a wide range of fields such as medical diagnosis, anomaly analysis and personalized advertising. The absence of any known negative labels makes it very challenging to learn binary classifiers from such data. Many state-of-the-art methods reformulate the original classification risk with individual risks over positive and unlabeled data, and explicitly minimize the risk of classifying unlabeled data as negative. This, however, usually leads to classifiers with a bias toward negative predictions, i.e., they tend to recognize most unlabeled data as negative. In this paper, we propose a label distribution alignment formulation for PU learning to alleviate this issue. Specifically, we align the distribution of predicted labels with the ground-truth, which is constant for a given class prior. In this way, the proportion of samples predicted as negative is explicitly controlled from a global perspective, and thus the bias toward negative predictions could be intrinsically eliminated. On top of this, we further introduce the idea of functional margins to enhance the model's discriminability, and derive a margin-based learning framework named Positive-Unlabeled learning with Label Distribution Alignment (PULDA). This framework is also combined with the class prior estimation process for practical scenarios, and theoretically supported by a generalization analysis. Moreover, a stochastic mini-batch optimization algorithm based on the exponential moving average strategy is tailored for this problem with a convergence guarantee. Finally, comprehensive empirical results demonstrate the effectiveness of the proposed method.

Multicategory Outcome Weighted Margin-based Learning for Estimating Individualized Treatment Rules.

Due to heterogeneity for many chronic diseases, precise personalized medicine, also known as precision medicine, has drawn increasing attentions in the scientific community. One main goal of precision medicine is to develop the most effective tailored therapy for each individual patient. To that end, one needs to incorporate individual characteristics to detect a proper individual treatment rule (ITR), by which suitable decisions on treatment assignments can be made to optimize patients' clinical outcome. For binary treatment settings, outcome weighted learning (OWL) and several of its variations have been proposed recently to estimate the ITR by optimizing the conditional expected outcome given patients' information. However, for multiple treatment scenarios, it remains unclear how to use OWL effectively. It can be shown that some direct extensions of OWL for multiple treatments, such as one-versus-one and one-versus-rest methods, can yield suboptimal performance. In this paper, we propose a new learning method, named Multicategory Outcome weighted Margin-based Learning (MOML), for estimating ITR with multiple treatments. Our proposed method is very general and covers OWL as a special case. We show Fisher consistency for the estimated ITR, and establish convergence rate properties. Variable selection using the sparse l 1 penalty is also considered. Analysis of simulated examples and a type 2 diabetes mellitus observational study are used to demonstrate competitive performance of the proposed method.

Integrated inference and learning of neural factors in structural support vector machines

Tackling pattern recognition problems in areas such as computer vision, bioinformatics, speech or text recognition is often done best by taking into account task-specific statistical relations between output variables. In structured prediction, this internal structure is used to predict multiple outputs simultaneously, leading to more accurate and coherent predictions. Structural support vector machines (SSVMs) are nonprobabilistic models that optimize a joint input–output function through margin-based learning. Because SSVMs generally disregard the interplay between unary and interaction factors during the training phase, final parameters are suboptimal. Moreover, its factors are often restricted to linear combinations of input features, limiting its generalization power. To improve prediction accuracy, this paper proposes: (i) joint inference and learning by integration of back-propagation and loss-augmented inference in SSVM subgradient descent; (ii) extending SSVM factors to neural networks that form highly nonlinear functions of input features. Image segmentation benchmark results demonstrate improvements over conventional SSVM training methods in terms of accuracy, highlighting the feasibility of end-to-end SSVM training with neural factors.

Towards a deeper geometric, analytic and algorithmic understanding of margins

Given a matrix A, a linear feasibility problem (of which linear classification is a special case) aims to find a solution to a primal problem or a certificate for the dual problem which is a probability distribution . Inspired by the continued importance of ‘large-margin classifiers’ in machine learning, this paper studies a condition measure of A called its margin that determines the difficulty of both the above problems. To aid geometrical intuition, we first establish new characterizations of the margin in terms of relevant balls, cones and hulls. Our second contribution is analytical, where we present generalizations of Gordan's theorem, and variants of Hoffman's theorems, both using margins. We end by proving some new results on a classical iterative scheme, the Perceptron, whose convergence rates famously depends on the margin. Our results are relevant for a deeper understanding of margin-based learning and proving convergence rates of iterative schemes, apart from providing a unifying perspective on this vast topic.

Confidence-weighted linear classification for text categorization

Confidence-weighted online learning is a generalization of margin-based learning of linear classifiers in which the margin constraint is replaced by a probabilistic constraint based on a distributi...

Nearest-neighbor classifier motivated marginal discriminant projections for face recognition

Marginal Fisher analysis (MFA) is a representative margin-based learning algorithm for face recognition. A major problem in MFA is how to select appropriate parameters, k 1 and k 2, to construct the respective intrinsic and penalty graphs. In this paper, we propose a novel method called nearest-neighbor (NN) classifier motivated marginal discriminant projections (NN-MDP). Motivated by the NN classifier, NN-MDP seeks a few projection vectors to prevent data samples from being wrongly categorized. Like MFA, NN-MDP can characterize the compactness and separability of samples simultaneously. Moreover, in contrast to MFA, NN-MDP can actively construct the intrinsic graph and penalty graph without unknown parameters. Experimental results on the ORL, Yale, and FERET face databases show that NN-MDP not only avoids the intractability, and high expense of neighborhood parameter selection, but is also more applicable to face recognition with NN classifier than other methods.

A Study on the Generalization Capability of Acoustic Models for Robust Speech Recognition

In this paper, we explore the generalization capability of acoustic model for improving speech recognition robustness against noise distortions. While generalization in statistical learning theory originally refers to the model's ability to generalize well on unseen testing data drawn from the same distribution as that of the training data, we show that good generalization capability is also desirable for mismatched cases. One way to obtain such general models is to use margin-based model training method, e.g., soft-margin estimation (SME), to enable some tolerance to acoustic mismatches without a detailed knowledge about the distortion mechanisms through enhancing margins between competing models. Experimental results on the Aurora-2 and Aurora-3 connected digit string recognition tasks demonstrate that, by improving the model's generalization capability through SME training, speech recognition performance can be significantly improved in both matched and low to medium mismatched testing cases with no language model constraints. Recognition results show that SME indeed performs better with than without mean and variance normalization, and therefore provides a complimentary benefit to conventional feature normalization techniques such that they can be combined to further improve the system performance. Although this study is focused on noisy speech recognition, we believe the proposed margin-based learning framework can be extended to dealing with different types of distortions and robustness issues in other machine learning applications.

A discriminative learning framework with pairwise constraints for video object classification

To deal with the problem of insufficient labeled data in video object classification, one solution is to utilize additional pairwise constraints that indicate the relationship between two examples, i.e., whether these examples belong to the same class or not. In this paper, we propose a discriminative learning approach which can incorporate pairwise constraints into a conventional margin-based learning framework. Different from previous work that usually attempts to learn better distance metrics or estimate the underlying data distribution, the proposed approach can directly model the decision boundary and, thus, require fewer model assumptions. Moreover, the proposed approach can handle both labeled data and pairwise constraints in a unified framework. In this work, we investigate two families of pairwise loss functions, namely, convex and nonconvex pairwise loss functions, and then derive three pairwise learning algorithms by plugging in the hinge loss and the logistic loss functions. The proposed learning algorithms were evaluated using a people identification task on two surveillance video data sets. The experiments demonstrated that the proposed pairwise learning algorithms considerably outperform the baseline classifiers using only labeled data and two other pairwise learning algorithms with the same amount of pairwise constraints.