Dexa/Daniel Bashir

Solving Multiclass Learning Problems

Thomas introduces the concept of error correcting output codes and its application in solving multiclass learning problems. Inspired by Terry Sanofsky's work on the netTALK system, he discusses the use of distributed representations and the analogy to error correct encodes.

In this clip
From this podcast
The Gradient
Thomas Dietterich: From the Foundations
Related Questions
- Is there research on error correction in learning?
- What are Type I and II errors in the episode 649: Introduction to Machine Learning — with Kirill Eremenko and Hadelin de Ponteves and the clip Understanding Model Errors?