4 ms·
At this point, all the computing power is concentrated among various companies such as Google, Facebook, Microsoft, Amazon, Tesla, etc. It seems to me it would
by ofou 2y ago
At this point, all the computing power is concentrated among various companies such as Google, Facebook, Microsoft, Amazon, Tesla, etc.
It seems to me it would be much safer and more intelligent to create a massive model and distribute the benefits among everyone. Why not use a P2P approach?
- nvy 2y agoIn my area, internet and energy are insanely expensive and that means I'm not at all willing to share my precious bandwidth or compute just to subsidize someone generating Rule 34 porn of their favorite anime character. I don't seed torrents for the same reason. If I lived in South Korea or somewhere that bandwidth was dirt cheap, then maybe.
- ofou 2y agoThere is a way to achieve load balancing, safety, and distribution effectively. The models used by Airbnb, Uber, and Spotify have proven to be generally successful. Peer-to-peer (P2P) technology is the future; even in China, people are streaming videos using this technology, and it works seamlessly. I envision a future where everyone joins the AI revolution with an iPhone, with both training and inference distributed in a P2P manner. I wonder why no one has done this yet.
- wizzwizz4 2y agoBackprop is neither commutative nor associative.
- ofou 2y agoWhat do you mean? There's a bunch of proof-of-concepts such as Hydra, peer-nnet, Learnae, and so on.
- wizzwizz4 2y agoThe Wikipedia article goes into more detail. https://en.wikipedia.org/wiki/Federated_learning https://en.wikipedia.org/wiki/Federated_learning
- whimsicalism 2y agojust because it has a PoC or a wiki article doesn’t mean it actually works