Sorry, you need to enable JavaScript to visit this website.

Compact Kernel Models for Acoustic Modeling via Random Feature Selection

Citation Author(s):
Michael Collins, Daniel Hsu, Brian Kingsbury
Submitted by:
Avner May
Last updated:
23 March 2016 - 7:16pm
Document Type:
Poster
Document Year:
2016
Event:
Presenters:
Avner May
Paper Code:
3484
 

A simple but effective method is proposed for learning compact random feature models that approximate non-linear kernel methods, in the context of acoustic modeling. The method is able to explore a large number of non-linear features while maintaining a compact model via feature selection more efficiently than existing approaches. For certain kernels, this random feature selection may be regarded as a means of non-linear feature selection at the level of the raw input features, which motivates additional methods for computational improvements. An empirical evaluation demonstrates the effectiveness of the proposed method relative to the natural baseline method for kernel approximation.

up
0 users have voted: