reinforcement learning is a decision making method with no known model of the environment at all.
- agent interacts with environment directly
- designer provide a performance measure of the agent in the environment
- agent tries to optimize the decision making algorithm to maximise the performance measure
Note: agent’s own choice of action, in this case, actually influences how the environment works (and what futures the agent sees). So the agent’s actions will influence the environment outcomes
contrast v. explicit programming v. planning
Note 2: look ma, no model! unlike optimization, reinforcement learning tasks does not require an optimization objective connected to a model of the environment where we know what knobs to turn. Instead, the objective is a literal performance of how the agent is doing in the actual environment.