video thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Spectral Learning of General Weighted Automata via Constrained Matrix Completion

Published on Feb 4, 20253534 Views

Many tasks in text and speech processing and computational biology involve functions from variable-length strings to real numbers. A wide class of such functions can be computed by weighted automat

Related categories

Presentation

Spectral Learning of General Weighted Automata via Constrained Matrix Completion00:00
The Problem { Regression over Strings00:16
Hypothesis Class: Weighted Finite Automata (WFA)02:05
Overview04:21
Spectral Learning in a Nutshell05:08
Spectral Learning by...06:39
The Hankel Matrix07:10
Spectral Learning for WFA09:01
Missing Entries10:24
Constrained Matrix Completion11:35
A Family of Algorithms13:34
Generalization Bound14:22
Want to Know More?15:59
Thank you!16:02