Mathematical Foundations for Machine Learning

Lecture notes

1. Introduction and error analysis

2. Optimization in Deep Learning

3. Deep Neural Networks and Dynamical Systems

4. Advanced Deep Neural Networks

1. Attention, Self-Attention, Transformer, Bert

5. Generative Deep Learning

1. VAE, GAN, NF

6. Deep Reinforcement

1. Dynamical Programing, Monte Carlo and TD Learning

7. Deep Reinforcement

1. Model Free RL - DDPG, PPO, TRPO

8. Deep Reinforcement

1. Model Based RL - RS, CEM, PETS, POPLIN, MBPO, M2AC, Latent space modeling

9. Software

10. Research Discussion and Presentations

References

• Ian GoodFellow, Yoshua Benjio, Aaron Conrville – Deep Learning

• Richard S. Sutton, Andrew G. Barto: Refincement Learning:An introduction

• Dimitri P. Bertsekas, reinforcement learning and optimal control