TY - GEN
T1 - Fast projections onto ℓ1,q-norm balls for grouped feature selection
AU - Sra, Suvrit
PY - 2011
Y1 - 2011
N2 - Joint sparsity is widely acknowledged as a powerful structural cue for performing feature selection in setups where variables are expected to demonstrate "grouped" behavior. Such grouped behavior is commonly modeled by Group-Lasso or Multitask Lasso-type problems, where feature selection is effected via ℓ1,q-mixed-norms. Several particular formulations for modeling groupwise sparsity have received substantial attention in the literature; and in some cases, efficient algorithms are also available. Surprisingly, for constrained formulations of fundamental importance (e.g., regression with an ℓ1,∞-norm constraint), highly scalable methods seem to be missing. We address this deficiency by presenting a method based on spectral projected-gradient (SPG) that can tackle ℓ1,q- constrained convex regression problems. The most crucial component of our method is an algorithm for projecting onto ℓ1,q-norm balls. We present several numerical results which show that our methods attain up to 30X speedups on large ℓ1,∞-multitask lasso problems. Even more dramatic are the gains for just the ℓ1,∞-projection subproblem: we observe almost three orders of magnitude speedups compared against the currently standard method.
AB - Joint sparsity is widely acknowledged as a powerful structural cue for performing feature selection in setups where variables are expected to demonstrate "grouped" behavior. Such grouped behavior is commonly modeled by Group-Lasso or Multitask Lasso-type problems, where feature selection is effected via ℓ1,q-mixed-norms. Several particular formulations for modeling groupwise sparsity have received substantial attention in the literature; and in some cases, efficient algorithms are also available. Surprisingly, for constrained formulations of fundamental importance (e.g., regression with an ℓ1,∞-norm constraint), highly scalable methods seem to be missing. We address this deficiency by presenting a method based on spectral projected-gradient (SPG) that can tackle ℓ1,q- constrained convex regression problems. The most crucial component of our method is an algorithm for projecting onto ℓ1,q-norm balls. We present several numerical results which show that our methods attain up to 30X speedups on large ℓ1,∞-multitask lasso problems. Even more dramatic are the gains for just the ℓ1,∞-projection subproblem: we observe almost three orders of magnitude speedups compared against the currently standard method.
UR - http://www.scopus.com/inward/record.url?scp=80052405487&partnerID=8YFLogxK
U2 - 10.1007/978-3-642-23808-6_20
DO - 10.1007/978-3-642-23808-6_20
M3 - Conference contribution
AN - SCOPUS:80052405487
SN - 9783642238079
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 305
EP - 317
BT - Machine Learning and Knowledge Discovery in Databases - European Conference, ECML PKDD 2011, Proceedings
T2 - European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, ECML PKDD 2011
Y2 - 5 September 2011 through 9 September 2011
ER -