Skip to content
Learn Motion
ExploreHow it worksMembership
Log in
Learn Motion

How AI Learns from Rewards and Mistakes

1The Trial-and-Error Loop: How an Agent Learns2Formalizing the Problem: States, Actions, Rewards, and Policies3Measuring Success: Return, Discounting, and Value4Learning from Experience: Value Updates and Q-Learning5Exploring vs. Exploiting: Where Mistakes Come From6Learning Policies Directly and Scaling Up
The Trial-and-Error Loop: How an Agent Learns

A Score, Not an Answer Key

2 / 2
Think about the game example. The agent presses a button, and three seconds later the score rises. The reward is genuine, but it arrives detached from the press that caused it. Compare that with the photo-labeling case: there, every image comes with the correct answer written on it, so the model knows exactly what it should have said. The reward gives no such answer. It only tells the agent that the sequence ended well, leaving the agent to figure out which of its recent choices deserves the credit.
0:00 / 0:00

A reward scores what happened. It never states which action was correct, so the agent must infer the connection between its decisions and the outcome.

Two kinds of learning signal

Supervised label

  • Supplies the correct output for each input
  • The model knows exactly what it should have predicted
  • Error is measured directly against the answer

Reinforcement reward

  • Supplies a scalar score for the outcome
  • The agent does not know which action earned the score
  • Credit must be assigned across recent decisions

Delayed score in a game

An agent presses a button at time one and sees no immediate change. Three seconds later the score rises by one. The reward is real, but it is attached to a moment far removed from the action that caused it. If the agent had pressed several buttons in between, it cannot tell from the reward alone which press mattered. A supervised label would have said, at the moment of the press, whether that press was correct. The reward says only that the overall sequence ended well.

This gap between outcome and cause is called the credit assignment problem. It is why reinforcement learning needs repeated experience rather than a single labeled example: only many trials, with varying actions and outcomes, let the agent separate the decisions that help from those that do not.

References

  1. [1]Reinforcement learning (Wikipedia)en.wikipedia.org
Previous2 / 2Next

Learn Motion

Generate a course. Learn it properly.

Operated by Wuhan Daoyin Technology Co., Ltd.

Contact: [email protected]
Privacy PolicyTerms of Service

© 2026 Learn Motion