(yet) (CS285) Lecture 19 - Reframing Control as an Inference Problem

이 글은 UC Berkeley 의 교수, Sergey Levine 의 심층 강화 학습 (Deep Reinforcement Learning) 강의인 CS285를 듣고 작성한 글 입니다.

Lecture 19의 강의 영상과 자료는 아래에서 확인하실 수 있습니다.


< 목차 >


Overview

slide1 Slide. 1.

slide2 Slide. 2.

Optimal Control as Model of Human Behavior

slide3 Slide. 3.

slide4 Slide. 4.

slide5 Slide. 5.

slide6 Slide. 6.

slide7 Slide. 7.

slide9 Slide. 9.

slide10 Slide. 10.

slide11 Slide. 11.

slide12 Slide. 12.

slide13 Slide. 13.

slide14 Slide. 14.

slide15 Slide. 15.

slide16 Slide. 16.

slide17 Slide. 17.

slide18 Slide. 18.

slide19 Slide. 19.

slide20 Slide. 20.

Control as Variational Inference

slide21 Slide. 21.

slide22 Slide. 22.

slide23 Slide. 23.

slide24 Slide. 24.

slide25 Slide. 25.

slide26 Slide. 26.

slide27 Slide. 27.

slide28 Slide. 28.

slide29 Slide. 29.

Algorithms for RL as Inference

slide30 Slide. 30.

slide31 Slide. 31.

slide32 Slide. 32.

slide33 Slide. 33.

slide34 Slide. 34.

slide35 Slide. 35.

Example Methods

slide36 Slide. 36.

slide37 Slide. 37.

slide38 Slide. 38.

slide39 Slide. 39.

slide40 Slide. 40.

slide41 Slide. 41.

slide42 Slide. 42.

slide43 Slide. 43.

slide44 Slide. 44.

slide45 Slide. 45.

asd

Reference