Fall-2021 Syllabus Zoom ID：984 9626 6594 and PW：753410
Videos: Aug24 Aug26 Aug31 Sep2 Sep7 Sep9 Sep14 Sep16 Sep21 Sep23 Sep28 Sep30 Oct 5 Oct 7 Oct 14 Oct 19 Oct 21 Oct 26 Oct 28 Nov 2 Nov 4 Nov 9 Nov 11 Nov 16 Nov 18 Nov 23
Notes: Lecture 1.1 Lecture 1.2 Lecture 1.3 Lecture 1.4 Lecture 1.5 Lecture 1.6 Lecture 2.1 Lecture 2.2 Lecture 3.1
Class project presentations:
Deep Reinforcement Learning on Imperfect Information games via Neural Fictitious Self-Play [video]
Is Reinforcement Learning More Difficult Than Bandits: A Near-optimal Algorithm Escaping the Curse of Horizon [video]
On the Susceptibility of EEG Classifiers to Adversarial Attacks [video]
Reinforcement Learning for Pixel-Wise Image Processing [video]
The use of Fused Lasso penalty to improve the patch-based segmentation [video]
Hamilton-Jacobi Reachability Analysis Based on Physics-constrained Deep Operator Network [video]
Web Traffic Prediction Across Multiple Sites via Spatial-Temporal Graph Neural Network [video]
Comparing Q-Learning to the Metropolis-Hastings algorithm for Combinatorial Optimization [video]
Exploring the Convergence of PPO and RUDDER Methods of Reinforcement Learning [video]
Model-Based Reinforcement Learning with Adversarial Training for Beer Recommendation [video]
Motor Synergy Development in High-Performing Deep Reinforcement Learning Algorithms [video]
Performance improvement for Low rank transition matrix in reinforcement learning setup [video]
Quasi-Value based Rainbow Implimentation Towards Mobile Robotic Pursuit [video]
Visual deep reinforcement learning for drone landing [video]
Visualising Loss Landscape for PINNs [video]
A Saddle-Point View for Constrained Reinforcement Learning with General Utilities and Zero Long-Term Constraint Violation [video]
Apply ADMM-based Optimization to Decentralized RL [video]
Auto-tuning Bayesian Filtering for Model Identification and Updating Using Reinforcement Learning [video]
Improving adversarial robustness with test-time randomness [video]
Bayesian Deep Learning: Theory, Motivations and Challenges [video]
Objective Adaption via Pontryagin Differentiable Programming [video]
Reinforcement Learning Tuned PID Controller via Input Convex Neural Networks [video]
Reduced Order Modeling of Forced Isotropic Turbulence using LSTM Neural Networks [video]
Fall-2020 Syllabus
Videos: 1 machine learning basic 2 DNN 3 CNN (1) 4 CNN (2) 5 RNN 6 regularization (1) 7 regularization (2) 8 regularization (3) 9 deep learning for solving PDEs 10 deep learning for system identification (1) 11 deep learning for system identification (2) 12 learning to do 13 DNN approximation (1) 14 DNN approximation (2) 15 DNN optimization and generalization
Spring-2020
This is a methods course for juniors in any branch of engineering and science, designed to follow MA 262. Basic techniques for solving systems of linear ordinary differential equations. Series solutions for second order equations, including Bessel functions, Laplace transform, Fourier series, numerical methods, separation of variables for partial differential equations and Sturm-Liouville theory.
Videos: Mar12, Mar 24, Mar 26, Mar 31, April2, April7, April9, April14, April16, April21, April28, April30Part1, April30Part2, April30Part3
Notes: nMar3toMar10, nMar12, nMar24, nMar26, nMar31, nApril2, nApril7, nApril9, nApril14, nApril16, nApril21, nApril28, April30Summary