4 ms·
I wasn't thinking of DeepRL, but more on the more classical side of things with approximators other than neural NNs; but what you describe makes sense.
by PartiallyTyped 3y ago
I wasn't thinking of DeepRL, but more on the more classical side of things with approximators other than neural NNs; but what you describe makes sense.
- bo1024 3y agoOn that side, reinforcement learning bleeds over into control theory, so you're partly right.