Geometric Analysis of Nonconvex Optimization Landscapes for Overcomplete Learning

Qing Qu, Yuexiang Zhai, Xiao Li, Yuqian Zhang, Zhihui Zhu

Keywords: dictionary learning, optimization, representation learning

Wed Session 4 (17:00-19:00 GMT) [Live QA] [Cal]
Wed Session 5 (20:00-22:00 GMT) [Live QA] [Cal]
Wednesday: Optimisation I

Abstract: Learning overcomplete representations finds many applications in machine learning and data analytics. In the past decade, despite the empirical success of heuristic methods, theoretical understandings and explanations of these algorithms are still far from satisfactory. In this work, we provide new theoretical insights for several important representation learning problems: learning (i) sparsely used overcomplete dictionaries and (ii) convolutional dictionaries. We formulate these problems as $\ell^4$-norm optimization problems over the sphere and study the geometric properties of their nonconvex optimization landscapes. For both problems, we show the nonconvex objective has benign (global) geometric structures, which enable the development of efficient optimization methods finding the target solutions. Finally, our theoretical results are justified by numerical simulations.

Similar Papers

Short and Sparse Deconvolution --- A Geometric Approach
Yenson Lau, Qing Qu, Han-Wen Kuo, Pengcheng Zhou, Yuqian Zhang, John Wright,
Transferring Optimality Across Data Distributions via Homotopy Methods
Matilde Gargiani, Andrea Zanelli, Quoc Tran Dinh, Moritz Diehl, Frank Hutter,