P.E.A.R.L.


Platform to Explore Algorithms in Reinforcement Learning

Episodes


Credits


Content Development - Dhruv Chadha, Shashwat Raghuvanshi

Website Development - Dev Kalra, Mohit Nathrani

Illustrations/Graphics - Manish Chandra, Tarun Kumar Singh

Special thanks to Dr. Aruna Tiwari and Dr. Kapil Ahuja for mentoring this project.

Episode 1 : Introduction


How can we make machines mimic the way humans learn, so they too become expert decision makers?

Episode 2 : Mathematical Treatment


Mathematical modelling of the human decision making process.

Chapter 3 : Evaluation Improvement Interplay


A very interesting phenomenon occurs while searching for the optimal decision making strategy.

Chapter 4 : Dynamic Programming


One of the first and foundational methods to solve the Bellman Equations. Limited in utility, but very essential conceptually.

Chapter 5 : Monte Carlo Methods


What to do when we don't have the complete knowledge of the environment and its transitions?

Chapter 6 : Temporal Difference Learning


Why to wait till the end of the episode?