3 ms·
It's a really tough question of when to pull the plug. Marvin Minsky wrote a whole book in 1969 'proving' that neural nets cannot, fundamentally, do anything in
by bernulli 5y ago
It's a really tough question of when to pull the plug. Marvin Minsky wrote a whole book in 1969 'proving' that neural nets cannot, fundamentally, do anything interesting - yet, here we are.
- otabdeveloper4 5y agoHe was mostly right. Neural nets are mostly an artistic tool for creating awesome generative art. They don't give any scientific or mathematical answers.
- civilized 5y ago> They don't give any scientific or mathematical answers. Dead wrong. Predicting protein folding has been a scientific holy grail for decades, and neural network-based AlphaFold is the leader. No non-neural approach comes within a country mile of its performance. They created a fold database of unprecedented coverage which is being used in research.
- bernulli 5y agoHe was decidedly not right. His claim was that neural networks are fundamentally not able to perform an XOR step, limiting any relevant computational capabilities. That claim was easily remedied later by slightly changing network architectures later on.