We consider a discrete optimization formulation for learning sparse classifiers, where the outcome depends upon a linear combination of a small subset of features. Recent work has shown that mixed integer programming (MIP) can be used to solve (to optimality) l(0)-regularized regression problems at scales much larger than what was conventionally considered possible. Despite their usefulness, MIP-based global optimization approaches are significantly slower than the relatively mature algorithms for l(1)-regularization and heuristics for nonconvex regularized problems. We aim to bridge this gap in computation times by developing new MIP-based algorithms for l(0)-regularized classification. We propose two classes of scalable algorithms: an exact algorithm that can handle p approximate to 50, 000 features in a few minutes, and approximate algorithms that can address instances with p approximate to 10(6) in times comparable to the fast l(1)-based algorithms. Our exact algorithm is based on the novel idea of integrality generation, which solves the original problem (with p binary variables) via a sequence of mixed integer programs that involve a small number of binary variables. Our approximate algorithms are based on coordinate descent and local combinatorial search. In addition, we present new estimation error bounds for a class of l(0)-regularized estimators. Experiments on real and synthetic data demonstrate that our approach leads to models with considerably improved statistical performance (especially variable selection) compared to competing methods.
机构:
Technion Israel Inst Technol, Fac Ind Engn & Management, IL-3200 Haifa, IsraelTechnion Israel Inst Technol, Fac Ind Engn & Management, IL-3200 Haifa, Israel
Beck, Amir
Tetruashvili, Luba
论文数: 0引用数: 0
h-index: 0
机构:
Technion Israel Inst Technol, Fac Ind Engn & Management, IL-3200 Haifa, IsraelTechnion Israel Inst Technol, Fac Ind Engn & Management, IL-3200 Haifa, Israel
机构:
Technion Israel Inst Technol, Fac Ind Engn & Management, IL-32000 Haifa, IsraelTechnion Israel Inst Technol, Fac Ind Engn & Management, IL-32000 Haifa, Israel
Beck, Amir
Eldar, Yonina C.
论文数: 0引用数: 0
h-index: 0
机构:
Technion Israel Inst Technol, Fac Elect Engn, IL-32000 Haifa, IsraelTechnion Israel Inst Technol, Fac Ind Engn & Management, IL-32000 Haifa, Israel
机构:
Technion Israel Inst Technol, Fac Ind Engn & Management, IL-3200 Haifa, IsraelTechnion Israel Inst Technol, Fac Ind Engn & Management, IL-3200 Haifa, Israel
Beck, Amir
Tetruashvili, Luba
论文数: 0引用数: 0
h-index: 0
机构:
Technion Israel Inst Technol, Fac Ind Engn & Management, IL-3200 Haifa, IsraelTechnion Israel Inst Technol, Fac Ind Engn & Management, IL-3200 Haifa, Israel
机构:
Technion Israel Inst Technol, Fac Ind Engn & Management, IL-32000 Haifa, IsraelTechnion Israel Inst Technol, Fac Ind Engn & Management, IL-32000 Haifa, Israel
Beck, Amir
Eldar, Yonina C.
论文数: 0引用数: 0
h-index: 0
机构:
Technion Israel Inst Technol, Fac Elect Engn, IL-32000 Haifa, IsraelTechnion Israel Inst Technol, Fac Ind Engn & Management, IL-32000 Haifa, Israel