2 ms·
I know RL is very hard. But we have about 1 TeraWord of text in current datasets and about 10 TeraWord could be scraped if we did a thorough job. RL is how lan
by visarga 4y ago
I know RL is very hard. But we have about 1 TeraWord of text in current datasets and about 10 TeraWord could be scraped if we did a thorough job. RL is how language models can generate more. By solving many problems and training on problem solving, AI can create its own data. It's the AlphaGo way - build your own data to surpass human level.
- PartiallyTyped 4y agoThis does not compute. To generate more, you need to interact with an environment, and you also need an objective function. If we could magically generate more textual data, then we have a language model already and don't need to train another language model. You can't bootstrap a model for language synthesis unless you give it access to the internet to interact with users, at which point ... you have a Tay [1] [1] https://en.wikipedia.org/wiki/Tay_(bot) https://en.wikipedia.org/wiki/Tay_(bot)