This preview shows page 1. Sign up to view the full content.
Unformatted text preview: negative
value for the data points marked -. Lyla finds that no matter what initial weights she uses, and how
and how much she trains the neural net, its error rate on the training data never goes below about
1/8. Why? 11 Part C (12 points)
Stephanie has been training some sigmoid neural nets, but was so busy celebrating the
election results last week that she forgot which one was which.
This page contains diagrams of six neural nets, labeled A through F. The next page shows
six classifier functions. Determine which classifier came from which neural net.
The six answers should all be unique, but you can give reasonable non-unique answers for
partial credit. 12 13 Tear off sheet. You need not hand this in. 6 M M 5 S Density 4 S S M 3 M S 2 A B 1 0
0 1 2 3 4 5 6 Hardness 14 Tear off sheet. You need not hand this in. 15 Tear off sheet....
View Full Document
This document was uploaded on 03/17/2014 for the course EECS 6.034 at MIT.
- Fall '10
- Artificial Intelligence