Ch. 2 of my free online book about neural nets and deep learning now available: http://neuralnetworksanddeeplearning.com/chap2.html Covers backpropagation in depth.
-
-
Replying to @michael_nielsen
@michael_nielsen I wonder if it's possible/nicer to explain single module (layer) of a net, and then build up to composition of them.2 replies 1 retweet 0 likes -
Replying to @karpathy
@michael_nielsen as in, every layer has input vector, weight vector, makes output vector with an f_w(x). At backprop computes local deltas1 reply 0 retweets 0 likes -
Replying to @michael_nielsen
@michael_nielsen thanks. I'm working on a demo that shows forward/backward in complete detail, in visualization, all numbers. Excited1 reply 0 retweets 0 likes -
Replying to @michael_nielsen
@michael_nielsen convnetjs and for network drawing trying canvas, svg or just html elements. Id love your input once it's fleshed out a bit!3 replies 0 retweets 0 likes
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.