Deep learning algorithms can fare poorly when the training dataset suffers from heavy class-imbalance but the testing criterion requires good generalization on less frequent classes. We design two novel methods to improve performance in such scenarios. First, we propose a theoretically-principled label-distribution-aware margin (LDAM) loss motivated by minimizing a margin-based generalization bound. This loss replaces the standard cross-entropy objective during training and can be applied with prior strategies for training with class-imbalance such as re-weighting or re-sampling. Second, we propose a simple, yet effective, training schedule that defers re-weighting until after the initial stage, allowing the model to learn an initial representation while avoiding some of the complications associated with re-weighting or re-sampling. We test our methods on several benchmark vision tasks including the real-world imbalanced dataset iNaturalist 2018. Our experiments show that either of these methods alone can already improve over existing techniques and their combination achieves even better performance gains.

With the explosive growth of e-commerce and the booming of e-payment, detecting online transaction fraud in real time has become increasingly important to Fintech business. To tackle this problem, we introduce the TitAnt, a transaction fraud detection system deployed in Ant Financial, one of the largest Fintech companies in the world. The system is able to predict online real-time transaction fraud in mere milliseconds. We present the problem definition, feature extraction, detection methods, implementation and deployment of the system, as well as empirical effectiveness. Extensive experiments have been conducted on large real-world transaction data to show the effectiveness and the efficiency of the proposed system.

We investigate the prediction capability of the orthogonal greedy algorithm (OGA) in high-dimensional regression models with dependent observations. The rates of convergence of the prediction error of OGA are obtained under a variety of sparsity conditions. To prevent OGA from overfitting, we introduce a high-dimensional Akaike's information criterion (HDAIC) to determine the number of OGA iterations. A key contribution of this work is to show that OGA, used in conjunction with HDAIC, can achieve the optimal convergence rate without knowledge of how sparse the underlying high-dimensional model is.

We study learning properties of accelerated gradient descent methods for linear least-squares in Hilbert spaces. We analyze the implicit regularization properties of Nesterov acceleration and a variant of heavy-ball in terms of corresponding learning error bounds. Our results show that acceleration can provides faster bias decay than gradient descent, but also suffers of a more unstable behavior. As a result acceleration cannot be in general expected to improve learning accuracy with respect to gradient descent, but rather to achieve the same accuracy with reduced computations. Our theoretical results are validated by numerical simulations. Our analysis is based on studying suitable polynomials induced by the accelerated dynamics and combining spectral techniques with concentration inequalities.

This article carries out a large dimensional analysis of standard regularized discriminant analysis classifiers designed on the assumption that data arise from a Gaussian mixture model with different means and covariances. The analysis relies on fundamental results from random matrix theory (RMT) when both the number of features and the cardinality of the training data within each class grow large at the same pace. Under mild assumptions, we show that the asymptotic classification error approaches a deterministic quantity that depends only on the means and covariances associated with each class as well as the problem dimensions. Such a result permits a better understanding of the performance of regularized discriminant analsysis, in practical large but finite dimensions, and can be used to determine and pre-estimate the optimal regularization parameter that minimizes the misclassification error probability. Despite being theoretically valid only for Gaussian data, our findings are shown to yield a high accuracy in predicting the performances achieved with real data sets drawn from the popular USPS data base, thereby making an interesting connection between theory and practice.

Data-driven approaches are becoming more common as problem-solving techniques in many areas of research and industry. In most cases, machine learning models are the key component of these solutions, but a solution involves multiple such models, along with significant levels of reasoning with the models' output and input. Current technologies do not make such techniques easy to use for application experts who are not fluent in machine learning nor for machine learning experts who aim at testing ideas and models on real-world data in the context of the overall AI system. We review key efforts made by various AI communities to provide languages for high-level abstractions over learning and reasoning techniques needed for designing complex AI systems. We classify the existing frameworks based on the type of techniques and the data and knowledge representations they use, provide a comparative study of the way they address the challenges of programming real-world applications, and highlight some shortcomings and future directions.

As electronically stored data grow in daily life, obtaining novel and relevant information becomes challenging in text mining. Thus people have sought statistical methods based on term frequency, matrix algebra, or topic modeling for text mining. Popular topic models have centered on one single text collection, which is deficient for comparative text analyses. We consider a setting where one can partition the corpus into subcollections. Each subcollection shares a common set of topics, but there exists relative variation in topic proportions among collections. Including any prior knowledge about the corpus (e.g. organization structure), we propose the compound latent Dirichlet allocation (cLDA) model, improving on previous work, encouraging generalizability, and depending less on user-input parameters. To identify the parameters of interest in cLDA, we study Markov chain Monte Carlo (MCMC) and variational inference approaches extensively, and suggest an efficient MCMC method. We evaluate cLDA qualitatively and quantitatively using both synthetic and real-world corpora. The usability study on some real-world corpora illustrates the superiority of cLDA to explore the underlying topics automatically but also model their connections and variations across multiple collections.

We consider open domain event extraction, the task of extracting unconstraint types of events from news clusters. A novel latent variable neural model is constructed, which is scalable to very large corpus. A dataset is collected and manually annotated, with task-specific evaluation metrics being designed. Results show that the proposed unsupervised model gives better performance compared to the state-of-the-art method for event schema induction.

The Black Box Variational Inference (Ranganath et al. [2014]) algorithm provides a universal method for Variational Inference, but taking advantage of special properties of the approximation family or of the target can improve the convergence speed significantly. For example, if the approximation family is a transformation family, such as a Gaussian, then switching to the reparameterization gradient (Kingma and Welling [2014]) often yields a major reduction in gradient variance. Ultimately, reducing the variance can reduce the computational cost and yield better approximations. We present a new method to extend the reparameterization trick to more general exponential families including the Wishart, Gamma, and Student distributions. Variational Inference with Numerical Derivatives (VIND) approximates the gradient with numerical derivatives and reduces its variance using a tight coupling of the approximation family. The resulting algorithm is simple to implement and can profit from widely known couplings. Our experiments confirm that VIND effectively decreases the gradient variance and therefore improves the posterior approximation in relevant cases. It thus provides an efficient yet simple Variational Inference method for computing non-Gaussian approximations.

Annotating large unlabeled datasets can be a major bottleneck for machine learning applications. We introduce a scheme for inferring labels of unlabeled data at a fraction of the cost of labeling the entire dataset. We refer to the scheme as greedy split and label exploration (GSAL). GSAL greedily queries an oracle (or human labeler) and partitions a dataset to find data subsets that have mostly the same label. GSAL can then infer labels by majority vote of the known labels in each subset. GSAL makes the decision to split or label from a subset by maximizing a lower bound on the expected number of correctly labeled examples. GSAL improves upon existing hierarchical labeling schemes by using supervised models to partition the data, therefore avoiding reliance on unsupervised clustering methods that may not accurately group data by label. We design GSAL with strategies to avoid bias that could be introduced through this adaptive partitioning. We evaluate GSAL on labeling of three datasets and find that it outperforms existing strategies for adaptive labeling.