Designing Societally Beneficial Reinforcement Learning Systems
BAIR
APRIL 29, 2022
For example, a thermostat turns on a furnace according to the current temperature measurement. In turn, these secondary effects could also influence the temperature which the thermostat monitors, leading to a longer timescale feedback loop. Control feedback gives an agent the ability to react to unforeseen events (e.g.
Let's personalize your content