Feature Clustering for Accelerating Parallel Coordinate Descent
Conference
·
OSTI ID:1111248
We demonstrate an approach for accelerating calculation of the regularization path for L1 sparse logistic regression problems. We show the benefit of feature clustering as a preconditioning step for parallel block-greedy coordinate descent algorithms.
- Research Organization:
- Pacific Northwest National Lab. (PNNL), Richland, WA (United States)
- Sponsoring Organization:
- USDOE
- DOE Contract Number:
- AC05-76RL01830
- OSTI ID:
- 1111248
- Report Number(s):
- PNNL-SA-88340; 400470000
- Resource Relation:
- Conference: Advances in Neural Information Processing Systems 25: 26th Annual Conference on Neural Information Processing Systems (NIPS 2012), December 3-6, 2012, Lake Tahoe, Nevada, 28-36
- Country of Publication:
- United States
- Language:
- English
Similar Records
GRAFTING: FAST, INCREMENTAL FEATURE SELECTION USING GRADIENT DESCENT IN FUNCTION SPACE
Avoiding Communication in Primal and Dual Block Coordinate Descent Methods
SU-D-204-01: A Methodology Based On Machine Learning and Quantum Clustering to Predict Lung SBRT Dosimetric Endpoints From Patient Specific Anatomic Features
Journal Article
·
Fri Jul 05 00:00:00 EDT 2002
· Journal of Machine Learning Research
·
OSTI ID:1111248
Avoiding Communication in Primal and Dual Block Coordinate Descent Methods
Journal Article
·
Thu Jan 17 00:00:00 EST 2019
· SIAM Journal on Scientific Computing
·
OSTI ID:1111248
+1 more
SU-D-204-01: A Methodology Based On Machine Learning and Quantum Clustering to Predict Lung SBRT Dosimetric Endpoints From Patient Specific Anatomic Features
Journal Article
·
Wed Jun 15 00:00:00 EDT 2016
· Medical Physics
·
OSTI ID:1111248
+4 more