We show how to read a human-interpretable algorithm from a neural network's weights, by studying the circuits formed by the connections between individual neurons: https://t.co/oX47rfEmqx pic.twitter.com/YPtrowyt5k
— OpenAI (@OpenAI) March 10, 2020
from Twitter https://twitter.com/OpenAI
March 10, 2020 at 08:36AM
via IFTTT
No comments:
Post a Comment