Accelerating block coordinate descent for nonnegative tensor factorization
Copyright policy )Accelerating block coordinate descent for nonnegative tensor factorization
AbstractThis paper is concerned with improving the empirical convergence speed of block‐coordinate descent algorithms for approximate nonnegative tensor factorization (NTF). We propose an extrapolation strategy in‐between block updates, referred to as heuristic extrapolation with restarts (HER). HER significantly accelerates the empirical convergence speed of most existing block‐coordinate algorithms for NTF, in particular for challenging computational scenarios, while requiring a negligible additional computational budget.
- French National Centre for Scientific Research France
- University of Southampton United Kingdom
- University of Rennes 1 France
- University of Mons Belgium
- French Institute for Research in Computer Science and Automation France
Sciences informatiques, FOS: Computer and information sciences, Computer Science - Machine Learning, Nesterov extrapolation, block-coordinate descent, Physique, chimie, mathématiques & sciences de la terre, nonconvex optimization, Ingénierie électrique & électronique, Machine Learning (stat.ML), Factorization of matrices, 510, Ingénierie, informatique & technologie, Machine Learning (cs.LG), Physical, chemical, mathematical & earth Sciences, Statistics - Machine Learning, Multilinear algebra, tensor calculus, FOS: Mathematics, Mathematics - Numerical Analysis, Mathematics - Optimization and Control, Electrical & electronics engineering, nonnegative tensor factorization, Numerical linear algebra, Numerical methods for low-rank matrix approximation; matrix compression, Numerical Analysis (math.NA), Computer science, [STAT.ML] Statistics [stat]/Machine Learning [stat.ML], 004, Engineering, computing & technology, Mathématiques, Optimization and Control (math.OC), Mathematics
Sciences informatiques, FOS: Computer and information sciences, Computer Science - Machine Learning, Nesterov extrapolation, block-coordinate descent, Physique, chimie, mathématiques & sciences de la terre, nonconvex optimization, Ingénierie électrique & électronique, Machine Learning (stat.ML), Factorization of matrices, 510, Ingénierie, informatique & technologie, Machine Learning (cs.LG), Physical, chemical, mathematical & earth Sciences, Statistics - Machine Learning, Multilinear algebra, tensor calculus, FOS: Mathematics, Mathematics - Numerical Analysis, Mathematics - Optimization and Control, Electrical & electronics engineering, nonnegative tensor factorization, Numerical linear algebra, Numerical methods for low-rank matrix approximation; matrix compression, Numerical Analysis (math.NA), Computer science, [STAT.ML] Statistics [stat]/Machine Learning [stat.ML], 004, Engineering, computing & technology, Mathématiques, Optimization and Control (math.OC), Mathematics
selected citations These citations are derived from selected sources.This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).5 popularity This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.Top 10% influence This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).Average impulse This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.Top 10%
Citations provided by BIP!Popularity provided by BIP!