Talk:UFLDL Tutorial
From Ufldl
for
Talk:UFLDL Tutorial
Jump to:
navigation
,
search
The first backprop assignment should be one doing classification with MNIST -- doing squared error with 1-hot label vectors. As an extension to that assignment, they should then modify the loss/gradient to do cross-entropy / likelihood.
Return to
Talk:UFLDL Tutorial
.
Views
Page
Discussion
View source
History
Personal tools
Log in
ufldl resources
UFLDL Tutorial
Recommended Readings
wiki
Main page
Recent changes
Random page
Help
Search
Toolbox
What links here
Related changes
Special pages