10-703 Deep RL | Schedule

Date	Lecture	Readings	Logistics
M 08/29	Lecture #1 : Introduction to Reinforcement and Representation Learning [ slides ]	S & B Textbook, Ch1 Smith & Gasser. The Development of Embodied Cognition - Six Lessons from Babies Dan Wolpert't talk The real reason for brains
W 08/31	Lecture #2 : Multi-armed Bandits [ slides ]	S & B Textbook, Ch2 2.1-2.7 Russo et al. A Tutorial on Thompson Sampling, Ch1-Ch4. Optional after Ch4
F 09/02	Recitation #1: Neural Nets, TensorFlow & Keras, OpenAI Gym, Bandits [ slides ]	G, B & C Textbook, Ch9, Ch10 Tensorflow tutorial notebook OpenAI Gym tutorial notebook PyTorch tutorial notebook The TensorFlow High Level (Keras) API
M 09/05	Labor Day - No Classes
W 09/07	Lecture #3 : Markov Decision Processes, Value Iteration, Policy Iteration [ slides ]	S & B Textbook, Ch3, Ch4 The Path perspective on Value Learning (blogpost)	HW1 out (tentative)
F 09/09	Recitation #2: Bandits, MDPs & HW1 [ slides ]
M 09/12	Lecture #4 : Monte Carlo Learning and Temporal Difference Learning [ slides \| slides 2 ]	S & B Textbook, Ch5, Ch6
W 09/14	Lecture #5 : Monte Carlo Learning and Temporal Difference Learning (Cont.) , Planning, Monte Carlo Tree search [ slides \| slides 2 ]	S & B Textbook, Ch5, Ch6 S & B Textbook, Ch8.11
F 09/16	Recitation #3: HW1 [ slides ]
M 09/19	Lecture #6 : Function approximation in prediction and control, Deep Q-learning [ slides ]	S & B Textbook, Ch6, Ch7 7.1-7.3 S & B Textbook, Ch8.11 S & B Textbook, Ch9.1-9.3, 9.6 and Ch10.1 Mnih et al. Playing Atari with Deep Reinforcement Learning Deep Learning for Real-Time Atari Game Play Using Offline Monte-Carlo Tree Search Planning
W 09/21	Lecture #7 : Policy gradients, REINFORCE, Actor-Critic methods [ slides ]	Mnih et al. Asynchronous Methods for Deep Reinforcement Learning S & B Textbook, Ch13 http://karpathy.github.io/2016/05/31/rl/
F 09/23	Recitation #4: MCTS, TD Learning, Deep Q Learning [ slides ]
M 09/26	Lecture #8 : Policy gradients, REINFORCE, Actor-Critic methods [ slides ]	Mnih et al. Asynchronous Methods for Deep Reinforcement Learning S & B Textbook, Ch13 http://karpathy.github.io/2016/05/31/rl/	HW1 due 11:59pm
W 09/28	Lecture #9 : Natural PG, PPO, TRPO [ slides ]	S & B Textbook, Ch13 Schulman et al. Trust Region Policy Optimization Schulman et al. Proximal Policy Optimization Algorithms (optional) Rajeswaran et al. Towards Generalization and Simplicity in Continuous Control	HW2 out (tentative)
F 09/30	Recitation #5: Quiz 1 Review [ slides ]
M 10/03	Lecture #10 : Maximum Entropy RL, soft actor critic, Deterministic Policy gradient, re-parametrized PG [ slides \| slides 2 ]	Lillicrap et al. Continuous control with deep reinforcement learning Haarnoja et al. Reinforcement Learning with Deep Energy-Based Policies Haarnoja et al. Soft Actor-Critic:Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor
W 10/05	Lecture #11 : Evolutionary methods for policy search [ slides ]	Salimans et al. Evolution Strategies as a Scalable Alternative to Reinforcement Learning (optional) Nikolaus Hansen. The CMA Evolution Strategy - A Tutorial, optional Antoine Cully et al. Robots that can adapt like animals
F 10/07	Quiz 1
M 10/10	Lecture #12 : Imitation learning, behavior cloning [ slides ]	Chen et al. Learning by Cheating Bagnell. An Invitation to Imitation, Up To Page 10 Bojarski et al. End to End Learning for Self-Driving Cars Andrychowicz et al. Hindsight Experience Replay
W 10/12	Lecture #13 : Adversarial imitation learning, Imitation learning for vision-based manipulation [ slides \| slides 2 ]	(optional) Ho et al. Generative Adversarial Imitation Learning Zhu et al. Reinforcement and Imitation Learning for Diverse Visuomotor Skills Zeng et. al Transporter Networks Rearranging the Visual World for Robotic Manipulation Seita et al. Learning to Rearrange Deformable Cables, Fabrics, and Bags with Goal-Conditioned Transporter Networks Ding et al. Goal-conditioned Imitation Learning
F 10/14	Recitation #6: Solutions to Quiz 1 [ slides ]
M 10/17	Fall Break - No Classes
W 10/19	Fall Break - No Classes
F 10/21	Fall Break - No Classes
M 10/24	Lecture #14 : Transporter networks for Robot manipulation (cont.), Model-based RL in low-dim state space [ slides \| slides 2 ]	Zeng et. al Transporter Networks Rearranging the Visual World for Robotic Manipulation Seita et al. Learning to Rearrange Deformable Cables, Fabrics, and Bags with Goal-Conditioned Transporter Networks Chua et al. Deep Reinforcement Learning in a Handful of Trials using Probabilistic Dynamics Models	HW3 out (tentative), HW2 due 11:59PM
W 10/26	Lecture #15 : MBRL (cont), AlphaGo, AlphaGoZero [ slides ]	(optional) David Silver et al. Mastering the game of Go with deep neural networks and tree search David Silver et al. Mastering the game of Go without human knowledge David Silver et al. Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm
F 10/28	Tartan Community Day - No Classes
M 10/31	Lecture #16 : AlphaGoZero, MBRL in sensory space [ slides \| slides 2 ]	Hansen et al. Temporal Difference Learning for Model Predictive Control Oh et al. Action-Conditional Video Prediction using Deep Networks in Atari Games (optional) Kaiser et al. Model-Based Reinforcement Learning for Atari
W 11/02	Lecture #17 : MBRL (cont.) Deterministic latent dynamics models [ slides \| slides 2 ]	DeepMind Blog Post MuZero:Mastering Go, chess, shogi and Atari without rules Julian Schrittwieser et al. Mastering Atari, Go, chess and shogi by planning with a learned model
F 11/04	Recitation #7: Quiz 2 Review [ slides ]
M 11/07	Lecture #18 : MBRL (cont.) Stochastic latent dynamics models [ slides \| slides 2 ]	Hafner, et.al Dream to Control:Learning Behaviors by Latent Imagination Hafner, et.al Mastering Atari with Discrete World Models Carl Doersch et al. Tutorial on Variational Autoencoders	HW4 out (tentative), HW3 due 11:59PM
W 11/09	Lecture #19 : Dynamics learning with graph neural networks [ slides \| slides 2 ]	Fragkiadaki et al. Learning Visual Predictive Models of Physics for Playing Billiards Gonzalez et al. Learning to Simulate Complex Physics with Graph Networks
F 11/11	Quiz 2
M 11/14	Lecture #20 : Intelligent Exploration [ slides ]	(optional) Osband et al. Deep Exploration via Bootstrapped DQN (optional) Pathak et al. Curiosity-driven Exploration by Self-supervised Prediction Burda et al. Large-Scale Study of Curiosity-Driven Learning Savinov et al. Episodic Curiosity through Reachability Ecoffet et al. Go-Explore: a New Approach for Hard-Exploration Problems Salimans et al. Learning Montezuma's Revenge from a Single Demonstration
W 11/16	Lecture #21 : Postponed [ slides ]
F 11/18	Recitation #8: Recitation [ slides ]
M 11/21	Lecture #22 : Offline RL, Learning by Observation [ slides ]	Fujimoto et al. Off-Policy Deep Reinforcement Learning without Exploration Mandlekar et al. IRIS Implicit Reinforcement without Interaction at Scale for Learning Control from Offline Robot Manipulation Data Hahn et al. No RL, No Simulation:Learning to Navigate without Navigating	HW5 out (tentative), HW4 due 11:59PM
W 11/23	Lecture #23 (Zoom Only): Sim2Real Transfer [ slides ]	Kumar et al. Rapid Motor Adaptation for Legged Robots Tobin et al. Domain Randomization for Transferring Deep Neural Networks from Simulation to the Real World Muller et al. Driving Policy Transfer via Modularity and Abstraction Akkaya et al. Solving Rubik’s Cube with A Robot Hand
F 11/25	Thanksgiving Break - No Classes
M 11/28	Lecture #24 : Visual Imitation Learning [ slides \| slides 2 ]	Hahn et al. No RL, No Simulation:Learning to Navigate without Navigating Aytar et al. Playing hard exploration games by watching YouTube Peng et al. SFV Reinforcement Learning of Physical Skills from Videos
W 11/30	Lecture #25 : Self-Supervised Visual Learning [ slides ]	Chen et al. A Simple Framework for Contrastive Learning of Visual Representations Srinivas et al. CURL Contrastive Unsupervised Representations for Reinforcement Learning Radford et al. Learning Transferable Visual Models From Natural Language Supervision Khandelwal et al. Simple but Effective CLIP Embeddings for Embodied AI
F 12/02	Recitation #9: Homework 5 [ slides ]
M 12/05	Lecture #26 (Katerina): Control with 3D visual representations [ slides ]	(optional) Fish Tung et al. Learning Spatial Common Sense with Geometry-Aware Recurrent Networks (optional) Fish Tung et al. 3D-OES Viewpoint-Invariant Object-Factorized Environment Simulators (optional) Harley et al. Learning from Unlabelled Videos Using Contrastive Predictive Neural 3D Mapping (optional) Yang et al. Visually-Grounded Library of Behaviors for Manipulating Diverse Objects across Diverse Configurations and Views
W 12/07	Lecture #27 : Language-guided Control [ slides ]	(optional) Shridhar et al. CLIPort What and Where Pathways for Robotic Manipulation	HW5 due 11:59PM
F 12/09	Recitation #10: Quiz 3 Review [ slides ]
M 12/12	Quiz 3, 8:30 am - 11:30 am, GHC 4401