PY 580 Machine Learning for Physicists. Fall 2025

 



This is the website for PY 580, Machine Learning for Physicists This website will be updated with HWs and suggested readings. The class will start by covering the first half of our review: A high-bias, low-variance introduction to Machine Learning for physicsts. The review can be downloaded from Physics Reports. The Jupyter Notebooks can be downloaded from Github. In the second half of the class, we will pivot to to more advanced/modern topics such as self-supervised learning, large language models, and diffusion models. The goal is to have a "Part II" of the ML review that covers these topics and updates the old review.


Installing Python:

The easiest way to install Python is using the Anaconda distribution: Anaconda.
Please also install Jupyter Notebooks and/or Jupyter Lab: Jupyter. Alternatively, you can run the notebooks on Google Colab. Students are encouraged to code with modern AI coding tools. You can also use the SCC cluster.


Useful Resources:

David Mackay's Information Theory, Inference, and Learning Algorithms.

Abu Mustafa's Online Course: Learning from Data

Michael Neilsen's Neural Networks and Deep Learning

Python Packages: Python scikit-learn Library , PyTorch , TensorFlow


Syllabus, Grading, and Course Information: pdf


Final Project Information: Here is a handout describing the final project


Weeks 1-2:

Reading: Chapter 1-4/ Notebook 1.
Notebooks:Google colab Notebook 1 with sliders (please save a copy locally) or github notebook (ask Gemni to update to modern Python 3 -- non-interactive).
Other Viewing + Readings: Lectures 1-3 from Learning from Data. Chapters 1-3 of Information Theory, Inference, and Learning Algorithms. IPython Cookbook fourth feature recipe: Introduction to Machine Learning in Python with scikit-learn.
Homework 1: Infinite data experiments for polynomial regression Due date Sept 16th.
Notes: No class 9/11, 9/23, 9/25. (choosing time for make-up in class on 9/10)

Weeks 3:

Reading: 9/16-9/18 Chapter 4 and 5.
Notebooks:Google colab Notebook on the surprise of double descent and Notebook 2 on gradient methods

Homework 2: Notebook 2 questions. Due date Sept 23rd.

Weeks 4:

Reading: 9/23-9/25 Chapter 6.

Notebooks Notebook 3+4 on linear regression methods

Homework 2: Notebook 3 questions. Due date Sept 30th.

Weeks 5:

Reading: 9/30-10/2 Chapter 7 and Chapter 9. Optional (not in class): Chapter 8

Notebooks Notebook 5-7 Logistic Regression

Homework 3: Notebook on classification in high-dimenesions (courtesy of Huan Souza). Python Notebook (download or just open in Colab). Due date Oct 9th. (Posted on Oct 5.)

Weeks 6:

Reading: 10/7-10/9 Chapter 9-11

Notebooks See notebooks for chapters

Homework 4: CIFAR 100 Challenge (Due Oct 17th)

Weeks 7:

Reading: 10/15 Chapter 12-13

Notebooks See notebooks for chapters

Homework 4: CIFAR 100 Challenge (Due Oct 17th)

Weeks 8:

Reading: 10/21-10/23 Chapter 13,14 15,17

Notebooks Intution for High Dimensions Notebook; Understanding UMAP

Homework:

Weeks 9:

Reading: 10/28-10/30 Chapter 14, 17

Notebooks

Homework:

Weeks 10:

Reading: 11/4-11/6 Chapter 17 (VAEs) + Paper on VQ-VAE and VQ-VAE 2 paper

Notebooks: Notebook 20: VAEs and Ising Model

Homework: Write your own VQ-VAE and train on CIFAR 10 Due:11/11.

Weeks 11:

Reading: 11/11, 11/13 (no lecture- Plan your final projects in class). Journal club: Diffusion models. Please read: Notebooks
Please play with very nice Huggin Face Notebooks here. It is part of the full HugginFace "Diffusion Course" which I hear is excellent for practical tips.

Homework: Turn in plan for final project 11-18.

Weeks 12-13

Reading: 11/18, 11/20,11/25 (Introducing the Transformer). Journal Club. Notebooks:HugginFace "LLM Course" which I hear is excellent for practical tips.

Weeks 14

Reading: 12/02 (NO CLASS),12/04, 12/7, 12/9 (LLM Advanced). Journal Club.