Hacker News Reader

many people seem to discount this thread because "Karpathy is one of the most successful PhD students in his field"

perhaps instead of discounting his experience... it would be better to take his advice

iamtraskon A Neural Network in 11 Lines of Python

Per your last comment, I'm thinking about changing this blog series to be a "blog" version of academic papers... starting with this one.

https://iamtrask.github.io/2016/02/25/deepminds-neural-stack...

iamtraskon Show HN: Sense2vec model trained on all 2015 Reddit comments

....now we know

iamtraskon Teaching Machines to Read and Comprehend – Deepmind [pdf]

This paper is extraordinary! How has it been 9 hours with only one point?

iamtraskon SEC approves 200+ million Americans as Angel Investors

pure democracy via capitalism... interesting. Are the masses smarter than the 1% is generous?

iamtraskon Distributing a Fully Connected Neural Network Across a Cluster

Precisely! I highly encourage checking out the slide-deck for a graphical representation.

For every node in every other layer, I colocate the edge on the same machine. In this way, when a group of, say, 10 nodes in layer 1 are each sending a weighted message to a single node in layer 2... they can pre-combine their messages (weighted sum) and send only that value over the network. This happens for every node in the second layer, reducing network i/o (this is the first optimization).

iamtraskon Distributing a Fully Connected Neural Network Across a Cluster

This also might help... here are some slides graphically showing how the distribution works. http://prezi.com/hdctecihctdr/?utm_campaign=share&utm_medium...

iamtraskon Distributing a Fully Connected Neural Network Across a Cluster

i apologize for the verbosity and thickness. Happy to answer questions though. :)

iamtraskon Tutorial: Apache Spark on GPU(s)

benchmarks like that coming soon :)