News for hackers. Stuff that matters.
Re:Introduction to Artificial Neural Networks – Part 1 by dave_sullivan on Saturday December 07, 2013 @03:13PM (#6867508)
2006 is commonly cited as the year when "deep learning" started becoming practical. Pretty sure it was Hinton's group, but they used greedy unsupervised pre-training to get a good initialization of the weights, then followed by supervised finetuning of said weights. That result kicked off a lot of renewed interest in NNs, which then led to using GPUs for a 40x speedup, which then led to many more impressive results (and they just keep coming). It turns out the unsupervised pre-training isn't even necessary, go figure...
Re:Introduction to Artificial Neural Networks – Part 1 by chmike on Sunday December 08, 2013 @04:09AM (#6869337)