3 ms·
This reminds me of George Hotz's Comma.ai end to end reinforcement learning approach vs Tesla's feature-engineering based approach described in an article I rea
by tejohnso 5y ago
This reminds me of George Hotz's Comma.ai end to end reinforcement learning approach vs Tesla's feature-engineering based approach described in an article I read.
Hotz feels that "not only will comma outpace Tesla, but that Tesla will eventually adopt comma’s method."[1]
[1]: https://return.life/2022/03/07/george-hotz-comma-ride-or-die https://return.life/2022/03/07/george-hotz-comma-ride-or-die
Previous discussion on the article: https://news.ycombinator.com/item?id=30738763 https://news.ycombinator.com/item?id=30738763
- fnbr 5y agoI think an end-to-end RL approach will eventually work- but _eventually_ could be in a really long time. It’s also a question of scale: even if Comma’s approach is fundamentally better, how much better is it? If Tesla has 1000x more cars, and their approach is 10x worse, they’ll still improve 100x faster than Comma.