An efficient algorithm for a class of fused Lasso problems

Jun Liu, Lei Yuan, Jieping Ye

Research output: Chapter in Book/Report/Conference proceedingConference contribution

130 Scopus citations

Abstract

The fused Lasso penalty enforces sparsity in both the coefficients and their successive differences, which is desirable for applications with features ordered in some meaningful way. The resulting problem is, however, challenging to solve, as the fused Lasso penalty is both non-smooth and non-separable. Existing algorithms have high computational complexity and do not scale to large-size problems. In this paper, we propose an Efficient Fused Lasso Algorithm (EFLA) for optimizing this class of problems. One key building block in the proposed EFLA is the Fused Lasso Signal Approximator (FLSA). To efficiently solve FLSA, we propose to reformulate it as the problem of finding an "appropriate" subgradient of the fused penalty at the minimizer, and develop a Subgradient Finding Algorithm (SFA). We further design a restart technique to accelerate the convergence of SFA, by exploiting the special "structures" of both the original and the reformulated FLSA problems. Our empirical evaluations show that, both SFA and EFLA significantly outperform existing solvers. We also demonstrate several applications of the fused Lasso.

Original languageEnglish (US)
Title of host publicationKDD'10 - Proceedings of the 16th ACM SIGKDD International Conference on Knowledge Discovery and Data
Pages323-332
Number of pages10
DOIs
StatePublished - Sep 7 2010
Event16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD-2010 - Washington, DC, United States
Duration: Jul 25 2010Jul 28 2010

Publication series

NameProceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

Other

Other16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD-2010
CountryUnited States
CityWashington, DC
Period7/25/107/28/10

Keywords

  • Fused Lasso
  • Restart
  • Subgradient
  • ℓ regularization

ASJC Scopus subject areas

  • Software
  • Information Systems

Cite this