- λͺ©μ : UC Berkeleyμ 2018λ λ CS294 κ°μ μλ£μ "νμ΄μ¬κ³Ό μΌλΌμ€λ‘ λ°°μ°λ κ°ννμ΅" μλ£λ₯Ό κΈ°λ°μΌλ‘ μ¬μΈ΅κ°ννμ΅μ λν΄ κ³΅λΆνκΈ°.
- κΈ°κ°: 2019λ 8μ ~ 2020λ 1μ (μ’ λ£μΌ λ―Ένμ )
- μ°Έμ¬μ: μμ μ, κΉμΉμ, κΉμΆ©ν¬, κΉνμ€, μ΄λμ, μ΄μ μ°, μ΄ν΄μ€, μ ν¨μ , μ΅μν, μ΅μ€κ·, μ΅μμ°, ν©νμ€
- κΈ°ν κ·Έλ£Ή: AI Robotics KR
- κ°μ 리뷰 : 3μΈ 1ν ꡬμ±, 1μ£Ό 1κ°μ© κ°μ 리뷰 λ°ν
- μ½λ© : κ°μ μ§λμ λ§μΆ°μ μμ μ½λ μ§ν
- 리뷰 λ°ν μλ£, μ§μ μλ΅ κ΄λ ¨λ λ΄μ©μ λΉμΌ λ°ν νμμ μ 리ν΄μ κΉνλΈμ μ λ‘λ
-
pytorch
- μ΅μν, μ΄λμ, μμ μ
- ν©νμ€, μ΄μ μ°, κΉμΆ©ν¬
-
tensorflow & keras
- μ΅μμ°, μ ν¨μ , κΉμΉμ
- μ΄ν΄μ€, μ΅μ€κ·, κΉνμ€
Deposit : 3λ§μ λ¬΄λ¨ κ²°μ(λΉμΌ μ·¨μ) μ λ§μ μ°¨κ°
- 10λΆ : 2000μ
- 30λΆ : 3000μ
- 1μκ° μ΄ν : 5000μ
| μ€ν°λ λ΄μ© | λ μ§μ μκ° | λ°νμ |
|---|---|---|
| Lecture 2: Supervised Learning and Imitation | 19/08/29 | μ΄ν΄μ€, μ΅μ€κ·, κΉνμ€ |
| Lecture 4: Reinforcement Learning Introduction | 19/09/05 | ν©νμ€, μ΄μ μ°, κΉμΆ©ν¬ |
| Lecture 5: Policy Gradients Introduction | 19/09/19 | μ΅μμ°, μ ν¨μ , κΉμΉμ |
| Lecture 6: Actor-Critic Introduction | 19/09/26 | μ΅μν, μ΄λμ, μμ μ |
| Lecture 7: Value Functions and Q-Learning | 19/10/03 | μ΄ν΄μ€, μ΅μ€κ·, κΉνμ€ |
| Lecture 8: Advanced Q-Learning Algorithms | 19/10/10 | ν©νμ€, μ΄μ μ°, κΉμΆ©ν¬ |
| Lecture 9: Advanced Policy Gradients | 19/10/17 | μ΅μμ°, μ ν¨μ , κΉμΉμ |
| Lecture 10: Optimal Control and Planning | 19/10/24 | μ΅μν, μ΄λμ, μμ μ |
| Lecture 11: Model-Based Reinforcement Learning | 19/10/31 | μ΄ν΄μ€, μ΅μ€κ·, κΉνμ€ |
| Lecture 12: Advanced Model Learning and Images | 19/11/07 | ν©νμ€, μ΄μ μ°, κΉμΆ©ν¬ |
| Lecture 13: Learning Policies by Imitating Other Policies | 19/11/14 | μ΅μμ°, μ ν¨μ , κΉμΉμ |
| Lecture 14: Probability and Variational Inference Primer | 19/11/21 | μ΅μν, μ΄λμ, μμ μ |
| Lecture 15: Connection between Inference and Control | 19/11/28 | μ΄ν΄μ€, μ΅μ€κ·, κΉνμ€ |
| Lecture 16: Inverse Reinforcement Learning | 19/12/05 | ν©νμ€, μ΄μ μ°, κΉμΆ©ν¬ |
| Lecture 17: Exploration: Part 1 | 19/12/12 | μ΅μμ°, μ ν¨μ , κΉμΉμ |
| Lecture 18: Exploration: Part 2 | 19/12/19 | μ΅μν, μ΄λμ, μμ μ |
| Lecture 19: Transfer Learning and Multi-Task Learning | 19/12/26 | μ΄ν΄μ€, μ΅μ€κ·, κΉνμ€ |
| Lecture 20: Meta-Learning | 19/01/02 | ν©νμ€, μ΄μ μ°, κΉμΆ©ν¬ |
| Lecture 21: Parallelism and RL System Design | 19/01/09 | μ΅μμ°, μ ν¨μ , κΉ |
| Lecture 22: Advanced Imitation Learning and Open Problems | 19/01/16 | μ΅μν, μ΄λμ, μμ μ |
- μ€νλΌμΈμΌλ‘ λ§€μ£Ό λͺ©μμΌ 19μ 30λΆ~ 21μ 30λΆμ μ§νλ©λλ€. (2019.08 ~ 2020.01)
- μ€ν°λ νμ: λ§€μ£Ό λμκ°λ©΄μ κ·Έ μ£Όμ μ‘°κ° μ€λΉν μλ£λ₯Ό μ°Έκ³ ν΄μ μ΄λ‘ κ³΅λΆ + μ€μ΅ + μ§μμλ΅ λ° ν λ‘
- μ΄λ‘ κ³΅λΆ μλ£:
- [Lecture Slides](http://rail.eecs.berkeley.edu/deeprlcourse-fa18/)
- μ§μ μλ΅: κ°μ λ΄μ©κ³Ό κ΄λ ¨ν΄ μλ‘ μ§λ¬Ένκ³ μ견μ 곡μ ν©λλ€.
