OpenAI beating pro Dota players, Deepmind beating professional Go players is amazing. DRL (Deep Reinforcement Learning) is the next hot shot and I sure want to know RL. This is exciting , here's the complete first lecture, this is going to be so much fun. Keeping the Honor Code, let's dive deep into Reinforcement Learning.

Book:

• Reinforcement Learning: An Introduction, Sutton and Barto, 2nd Edition.
• Reinforcement Learning: State-of-the-Art, Marco Wiering and Martijn van Otterlo, Eds.
• Linear Algebra Review | Probability Review

Grade : Assignment 1 (10%) + Assignment 2 (20%) + Assignment 3 (15%) + Midterm (25%) + Quiz(5%) + Final Proejct (25%)

To realize the dreams and impact of AI requires autonomous systems that learn to make good decisions. Reinforcement learning is one powerful paradigm for doing so, and it is relevant to an enormous range of tasks, including robotics, game playing, consumer modeling and healthcare. This class will provide a solid introduction to the field of reinforcement learning and students will learn about the core challenges and approaches, including generalization and exploration. Through a combination of lectures, and written and coding assignments, students will become well versed in key ideas and techniques for RL. Assignments will include the basics of reinforcement learning as well as deep reinforcement learning — an extremely promising new area that combines deep learning techniques with reinforcement learning. In addition, students will advance their understanding and the field of RL through an open ended project.

• Homework 1 : Solution | Starter Code
• Homework 2 : Solution | Starter Code
• Homework 3 : Solution | Starter Code

• Introduction to Reinforcement Learning [Slide] SBC 1
• How to act given know how the world works. Tabular setting. Markov processes. Policy search. Policy iteration. Value iteration [Slide] SBC 3, 4.1-4.4
• Learning to evaluate a policy when don't know how the world works. [Slide] SBC 5.1, 5.5, 6.1-6.3
• Model-free learning to make good decisions. Q-learning. SARSA., Model Free Control [Slide] SBC 5.2, 5.4, 6.4-6.5, 6.7
• Scaling up: value function approximation. Deep Q Learning, Value Function Approximation [Slide] SBC Chp 9.3, 9.6-9.7, 10.1, 11.1, 11.2, 11.3
• Deep reinforcement learning continued [Slide] SBC 9.7
• Imitation Learning [Slide]
• Policy Search, Policy Gradient [Slide] SBC 13
• Fast reinforcement learning (Exploration/Exploitation) Part I [Slide] SBC Sections 2.1-2.7
• Fast reinforcement learning (Exploration/Exploitation) Part II [Slide] SBC Sections 2.1-2.7
• Batch Reinforcement Learning [Slide]
• Monte Carlo Tree Search [Slide]
• Human in the loop RL with a focus on transfer learnign [Slide]

The following guest lecture slides from last year's class offering may also help you in generating good project ideas.

• Cooperative Inverse Reinforcement Learning, Dylan Hadfield-Menell. Slides
• Maximum Entropy Framework: Inverse RL, Soft Optimality, and More, Chelsea Finn and Sergey Levine. Slides
• Reinforcement Learning – Policy Optimization Pieter Abbeel. Slides
• Safe Reinforcement Learning, Philip S. Thomas. Slides

• Human-level control through deep reinforcement learning
• Maximum Entropy Inverse Reinforcement Learning
• Apprenticeship Learning via Inverse Reinforcement Learning
• Monte-Carlo tree search and rapid action value estimation in computer Go
• Mastering the game of Go without human knowledge

• ICLR Result Reproducibility Challenge | Python Tutorial | OpenAI | Deep Reinforcement Learning | An introduction to Reinforcement Learning | Deep Reinforcement Learning Udacity | DeepMind : Advanced Deep Learning & Reinforcement | DeepMind : Introduction to Reinforcement Learning | DeepMind | awesome-rl

• Stanford Seminar - Human in the Loop Reinforcement Learning | CS231n Lecture 14 | Deep Reinforcement Learning | 60 Days of RL Challenge | Udacity DRL Nanodegree

The list of exciting project in RL domain goes on. Here are some of the project posters from past year and the ICML template to be followed. Good projects are encouraged to be in the standards of RLDM, AAMAS etc. Deep Reinforcement Learning, is exciting. Here is my project called, " ".