Trading convexity for scalability 论文

2006引用 363
DOIPDF
Machine Learning and AlgorithmsDomain Adaptation and Few-Shot LearningStochastic Gradient Optimization Techniques
人工智能Domain Adaptation and Few-Shot LearningStochastic Gradient Optimization TechniquesMachine Learning and Algorithms
相关技术:Stochastic Gradient Optimization TechniquesMachine Learning and AlgorithmsDomain Adaptation and Few-Shot Learning
关系图谱

4

作者

3

相关技术

0

相关事件

0

相关文章

摘要

Convex learning algorithms, such as Support Vector Machines (SVMs), are often seen as highly desirable because they offer strong practical properties and are amenable to theoretical analysis. However, in this work we show how non-convexity can provide scalability advantages over convexity. We show how concave-convex programming can be applied to produce (i) faster SVMs where training errors are no longer support vectors, and (ii) much faster Transductive SVMs.

作者查看全部 (4)

L

Léon Bottou

J

Jason Weston

F

Fabian H. Sinz

R

Ronan Collobert

相关技术查看全部 (3)

Stochastic Gradient Optimization TechniquesMachine Learning and AlgorithmsDomain Adaptation and Few-Shot Learning

相关事件

暂无数据

相关文章

暂无数据