• 목록

Course Description:

     · Introduces the design of algorithms for Reinforcement learning (RL) that enable machines to learn based  on reinforcements. 

     · RL allows machines to learn from partial, implicit and delayed feedback. 

     · RL is useful in sequential decision making tasks where a machine repeatedly interacts with the environment or users.

Course Overview:

Topics

     · Markov decision processes

     · Model free reinforcement learning

     · Model based reinforcement learning

     · Partially observable reinforcement learning

     · Deep reinforcement learning

     · Hierarchical reinforcement learning

     · Imitation learning

     · Inverse reinforcement learning

     · Meta learning

Potential applications

     · Robotic control · Game playing

     · Conversational agents

     · Operations research

     · Assistive technologies

     · Intelligent tutoring systems

     · Computational finance

     · Autonomous vehicles

     · Optimization