Plans for the week of March 24-28
Dear all, welcome back to FYS5429/9429. We hope you've had a great weekend.
This week we start with generative models, after a short reminder on autoencoders and a discussion on how to implements these using Tensorflow/Keras or Pytorch. We will spend the rest of the semester on these methods (and we get time we will end with a discussion on Transformers/Reinforcement learning).
Our plan is to cover energy based models, GANS, variational autoenconders and diffusion models. We will start with the basic math of these models, including Monte Carlo sampling. The plan for this week is thus to
Plans for the week March 24-28
-
Finalizing discussion on autoencoders and implementing Autoencoders with TensorFlow/Keras and PyTorch
-
Overview of generative models
-
Probability distributions and Markov Chain Monte Carlo simulations (Metropolis and Gibbs sampling)
-
If we get time, we end the lecture with Boltzmann machines and energy models
-
Reading recommendation: Goodfellow et al chapters 16 and 18.1 and 18.2. Chapter 17 gives a background to Monte Carlo Markov Chains.
- Short discussions for topics for project 2 (for those who plan to hand in two projects) during the lab session
The jupyter-notebook is at https://github.com/CompPhysics/AdvancedMachineLearning/blob/main/doc/pub/week10/ipynb/week10.ipynb
Best wishes to you all,
Edvin and Morten