LMAO. and a good example how statistics ≠ understanding.pic.twitter.com/BhKnSaYKkG
Show this thread
-
For me deep learning is a useful tool, just like computers are useful tools. We don't want to throw away all our computers (and maybe start from scratch to build artificial bio-beings) just because they haven't solved AGI yet or even matched the intelligence of 5 year old yet.
no, certainly not, and nobody is advocating that but a major figure in the field recently attacked me for claiming that certain neural nets would face problems in understanding novel sentences. in my view, the problems in understanding are quite real; i want to document them.
