4 ms·
All learning is reward-based. We can imitate human reward systems to start. The machines will act like humans do: basically random behaviors at the start, then
by douglaswlance 9y ago
All learning is reward-based. We can imitate human reward systems to start. The machines will act like humans do: basically random behaviors at the start, then following pleasure and rewards to shape their behavior.
- zi5yaolo8o6 9y agowhat is the reward for learning for the joy of learning ?
- oliv__ 9y agohmm.. joy?