A reward signal maps outcomes to a value used for learning or selection. If it is only a proxy for the real goal, optimization may exploit the proxy and harm unmeasured outcomes.

Reward Signals Shape Learned Behavior

A policy maps state to action, and reinforcement learning adjusts a policy from reward-bearing experience. Safety constraints and off-policy evaluation matter before learned choices control real tools.

Agent Graphs as Policies Over States

In this situation—rewarding a tutor for session length instead of durable learning—which choice best applies “Reward Signals Shape Learned Behavior”?

MCTS grows a search tree by balancing promising actions with underexplored ones and using simulated outcomes. Its result depends on rollout fidelity, budget, and value estimates.

Learn Before

Related