1 / 14

Cpt_S 580-03 Intelligent Agents

Cpt_S 580-03 Intelligent Agents. (Reinforcement Learning). Actions: Wave, Stand, Clap Observations: Colors, Rewards Goal: find an optimal policy. What is your policy?. What does the world look like?. Formalize the Problem. Knowns Unknowns. (UP) Wave (RIGHT) Stand (DOWN) Clap.

orinda
Télécharger la présentation

Cpt_S 580-03 Intelligent Agents

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Cpt_S 580-03Intelligent Agents (Reinforcement Learning)

  2. Actions: Wave, Stand, Clap • Observations: Colors, Rewards • Goal: find an optimal policy

  3. What is your policy?

  4. What does the world look like?

  5. Formalize the Problem • Knowns • Unknowns

  6. (UP) Wave • (RIGHT) Stand • (DOWN) Clap

  7. Agent can change policy • Representation and Reward as given • Pros: • Just specify goals • can be much less work than programming • unexpected situations • Cons: • Can be slow • need to pick representation & rewards

  8. Goals for Our Class • Solid foundation for RL • More focus on empirical than theoretical

  9. Introductions

  10. Reduced Formalism • Knowns • Unknowns

  11. Observations vs. State • Types of ML: Supervised, Unsupervised, RL • Value-based vs. Policy Search • Dynamic Programming vs. Model Free vs. Model-based (Model-learning) • Explore vs. Exploit

More Related