• Supervised Learning

  • Data : (x, y)

    x is data, y is label

  • Goal : Learn a function to map $x \to y$

  • Examples : Classification, regression, object detection, semantic segmentation, image captioning, etc.

• Unsupervised Learning

  • Data : x

    just data, no label

  • Goal : Learn some underlying hidden structure of the data

  • Examples : Clustering, dimensionality reduction, feature learning, density estimation, etc.

• Reinforcement Learning

  • Problems involving an agent interacting with an environment, which provides numeric reward signals

  • Goal : Learn how to take actions in order to maximize reward

    action과 reward에 대한 정의가 필요

  

What is Reinforcement Learning

Reinforcement Learning

Agent : 어떤 action을 취할 수 있는 주체

Environment : agent에 영향을 주고 상호작용할 수 있는 주변 환경, 조건

environment는 agent에게 t 시점에서의 상태 state $s_t$를 제공

agent는 t 시점에서의 action $a_t$를 취함

environment에 따라 reward $r_t$를 얻게 됨

그에 따라 다음 시점 t + 1 에서의 상태 $s_{t+1}$를 얻게 됨

Cart-Pole Problem