4 ms·
Reinforcement learning without simulation is not feasible for complex problems. If you can’t simulate, stick to multi-armed bandits.
by orasis 7y ago
Reinforcement learning without simulation is not feasible for complex problems.
If you can’t simulate, stick to multi-armed bandits.