Learn More
Motivated principally by the low-rank matrix completion problem, we present an extension of the Frank-Wolfe method that is designed to induce near-optimal solutions on low-dimensional faces of the feasible region. This is accomplished by a new approach to generating " in-face " directions at each iteration, as well as through new choice rules for selecting(More)
We present new results for the Frank-Wolfe method (also known as the conditional gradient method). We derive computational guarantees for arbitrary step-size sequences, which are then applied to various step-size rules, including simple averaging and constant step-sizes. We also develop step-size rules and computational guarantees that depend naturally on(More)
We present new results for the conditional gradient method (also known as the Frank-Wolfe method). We derive computational guarantees for arbitrary step-size sequences, which are then applied to various step-size rules, including simple averaging and constant step-sizes. We also develop step-size rules and computational guarantees that depend naturally on(More)
In this paper we analyze boosting algorithms [15, 21, 24] in linear regression from a new perspective: that of modern first-order methods in convex optimization. We show that classic boosting algorithms in linear regression, namely the incremental forward stagewise algorithm (FS ε) and least squares boosting (LS-Boost(ε)), can be viewed as subgradient(More)
Boosting methods are highly popular and effective supervised learning methods which combine weak learners into a single accurate model with good statistical performance. In this paper, we analyze two well-known boosting methods, AdaBoost and Incremental Forward Stagewise Regression (FS ε), by establishing their precise connections to the Mirror Descent(More)
We develop an optimization model and corresponding algorithm for the management of a demand-side platform (DSP), whereby the DSP aims to maximize its own pro€t while acquiring valuable impressions for its advertiser clients. We formulate the problem of pro€t maximization for a DSP interacting with ad exchanges in a real-time bidding environment in a(More)
and Objectives: Action learning seminar on analytics, machine learning and the digital economy The unprecedented growth in big data and analytics is driving a revolution in management decision-­‐‑ making, operations, marketing, finance, and product innovation. Businesses across the world are wrestling with challenges and opportunities that call for the(More)
  • 1