Online optimization has been a successful framework for solving large-scale problems under computational constraints and partial information. Current methods for online convex optimization require either a projection or exact gradient computation at each step, both of which can be prohibitively expensive for large-scale applications. At the same time, there is a growing trend of non-convex optimization in machine learning community and a need for online methods. Continuous submodular functions, which exhibit a natural diminishing returns condition, have recently been proposed as a broad class of non-convex functions which may be efficiently optimized. Although online methods have been introduced, they suffer from similar problems. In this work, we propose Meta-Frank-Wolfe, the first online projectionfree algorithm that uses stochastic gradient estimates. The algorithm relies on a careful sampling of gradients in each round and achieves the optimal $O(\sqrt{T})$ adversarial regret bounds for convex and continuous submodular optimization. We also propose One-Shot Frank-Wolfe, a simpler algorithm which requires only a single stochastic gradient estimate in each round and achieves a $O(T^{2/3})$ stochastic regret bound for convex and continuous submodular optimization. We apply our methods to develop a novel "lifting" framework for the online discrete submodular maximization and also see that they outperform current state of the art techniques on an extensive set of experiments.

This is the second Simpliv class on Matlab I've taken. Already, a couple important concepts have been discussed that weren't discussed in the previous course. I'm glad the instructor is comparing Matlab to Excel, which is the tool I've been using and have been frustrated with. This course is a little more advanced than the previous course I took. As an engineer, I'm delighted it covers complex numbers, derivatives, and integrals.

In the last blog post of this series, we discussed classifiers. The categories of classifiers and how they are evaluated were discussed. We have also discussed regression models in depth. In this post, we dwell a little deeper in how regression models can be used for classification tasks. Logistic Regression is a widely used regression model used for classification tasks.

Machine learning is an application of artificial intelligence that gives a system an ability to automatically learn and improve from experiences without being explicitly programmed. In this article, we have listed some of the best free machine learning books that you should consider going through (no order in particular). Based on the Stanford Computer Science course CS246 and CS35A, this book is aimed for Computer Science undergraduates, demanding no pre-requisites. This book has been published by Cambridge University Press. This book holds the prologue to statistical learning methods along with a number of R labs included.

Vote-boosting is a sequential ensemble learning method in which the individual classifiers are built on different weighted versions of the training data. To build a new classifier, the weight of each training instance is determined in terms of the degree of disagreement among the current ensemble predictions for that instance. For low class-label noise levels, especially when simple base learners are used, emphasis should be made on instances for which the disagreement rate is high. When more flexible classifiers are used and as the noise level increases, the emphasis on these uncertain instances should be reduced. In fact, at sufficiently high levels of class-label noise, the focus should be on instances on which the ensemble classifiers agree. The optimal type of emphasis can be automatically determined using cross-validation. An extensive empirical analysis using the beta distribution as emphasis function illustrates that vote-boosting is an effective method to generate ensembles that are both accurate and robust.

Actor-critic methods solve reinforcement learning problems by updating a parameterized policy known as an actor in a direction that increases an estimate of the expected return known as a critic. However, existing actor-critic methods only use values or gradients of the critic to update the policy parameter. In this paper, we propose a novel actor-critic method called the guide actor-critic (GAC). GAC firstly learns a guide actor that locally maximizes the critic and then it updates the policy parameter based on the guide actor by supervised learning. Our main theoretical contributions are two folds. First, we show that GAC updates the guide actor by performing second-order optimization in the action space where the curvature matrix is based on the Hessians of the critic. Second, we show that the deterministic policy gradient method is a special case of GAC when the Hessians are ignored. Through experiments, we show that our method is a promising reinforcement learning method for continuous controls.

Motivation: Thanks to digitization, we often have access to large databases, consisting of various fields of information, ranging from numbers to texts and even boolean values. Such databases lend themselves especially well to machine learning, classification and big data analysis tasks. We are able to train classifiers, using already existing data and use them for predicting the values of a certain field, given that we have information regarding the other fields. Most specifically, in this study, we look at the Electronic Health Records (EHRs) that are compiled by hospitals. These EHRs are convenient means of accessing data of individual patients, but there processing as a whole still remains a task. However, EHRs that are composed of coherent, well-tabulated structures lend themselves quite well to the application to machine language, via the usage of classifiers. In this study, we look at a Blood Transfusion Service Center Data Set (Data taken from the Blood Transfusion Service Center in Hsin-Chu City in Taiwan). We used scikit-learn machine learning in python. From Support Vector Machines(SVM), we use Support Vector Classification(SVC), from the linear model we import Perceptron. We also used the K.neighborsclassifier and the decision tree classifiers. We segmented the database into the 2 parts. Using the first, we trained the classifiers and the next part was used to verify if the classifier prediction matched that of the actual values. Contact: ritabratamaiti@hiretrex.com

We present a novel model architecture which leverages deep learning tools to perform exact Bayesian inference on sets of high dimensional, complex observations. Our model is provably exchangeable, meaning that the joint distribution over observations is invariant under permutation: this property lies at the heart of Bayesian inference. The model does not require variational approximations to train, and new samples can be generated conditional on previous samples, with cost linear in the size of the conditioning set. The advantages of our architecture are demonstrated on learning tasks requiring generalisation from short observed sequences while modelling sequence variability, such as conditional image generation, few-shot learning, set completion, and anomaly detection.

How does coarsening affect the spectrum of a general graph? We provide conditions such that the principal eigenvalues and eigenspaces of a coarsened and original graph Laplacian matrices are close. The achieved approximation is shown to depend on standard graph-theoretic properties, such as the degree and eigenvalue distributions, as well as on the ratio between the coarsened and actual graph sizes. Our results carry implications for learning methods that utilize coarsening. For the particular case of spectral clustering, they imply that coarse eigenvectors can be used to derive good quality assignments even without refinement---this phenomenon was previously observed, but lacked formal justification.

A nonparametric Bayesian sparse graph linear dynamical system (SGLDS) is proposed to model sequentially observed multivariate data. SGLDS uses the Bernoulli-Poisson link together with a gamma process to generate an infinite dimensional sparse random graph to model state transitions. Depending on the sparsity pattern of the corresponding row and column of the graph affinity matrix, a latent state of SGLDS can be categorized as either a non-dynamic state or a dynamic one. A normal-gamma construction is used to shrink the energy captured by the non-dynamic states, while the dynamic states can be further categorized into live, absorbing, or noise-injection states, which capture different types of dynamical components of the underlying time series. The state-of-the-art performance of SGLDS is demonstrated with experiments on both synthetic and real data.