Machine Learning in Physics

Lectures and tutorials to learn how to write machine learning programs with Python

Launch Tool

You must login before you can run this tool.

Version 1.0 - published on 04 Nov 2021

doi:10.21981/QADE-N936 cite this

Open source: license | download

View All Supporting Documents

Category

Tools

Published on

Abstract

This course introduces the fundamentals of machine learning as applied to problems in physical science.
The content was taken from AP40012 spring semester 2020 at Hong Kong Polytechnic University.
The course includes video lectures (+ slides), and interactive (Jupyter notebooks) lab sessions.
Topics covered in the course include:

  • Linear regression and classification: databases, feature, representation, training set, error functions, gradient descent, univariate and multi-variate regression, logistic regression;
  • Neural networks: activation functions, back propagation, feed-forward networks, recurrent networks, deep learning;
  • Theory of machine learning: overfitting, validation, regularization, generalization, bias-variance;
  • Application in physical science: Materials Project database, bulk modulus, bandgap, energy, force, molecular dynamics.

In this course students will learn how to program in Python with Jupyter notebooks and use external libraries such as Numpy, Scipy, Pandas and Tensorflow. Lectures introduce concepts including some illustrative examples and algorithms related to solving problems in physics or engineering. During computer laboratory students will write small programs to gain a deeper understanding of the topics discussed during the lectures.

Powered by

Python and Jupyter notebooks.

Bio

Nicolas Onofrio, Assistant Prof. at the Hong Kong Polytechnic University.

References

  • Hands-on machine learning with Scikit-Learn and TensorFlow: concepts, tools, and techniques to build intelligent systems. Géron, Aurélien. O'Reilly Media, Inc., 2017.
  • A first course in machine learning. Rogers, Simon, and Mark Girolami. Chapman and Hall/CRC, 2016.
  • Machine Learning, MOOC from Stanford (CS229). Ng, Andrew. 
  • Learning From Data, MOOC from Caltech. Yaser S. Abu-Mostafa.
  • Neural Networks and Deep Learning,  http://neuralnetworksanddeeplearning.com, Michael Nielsen

Cite this work

Researchers should cite this work as follows:

  • Please cite using DOI if you decide to use part of the codes in your publications.

  • Nicolas Onofrio (2021), "Machine Learning in Physics," https://nanohub.org/resources/mlphysics. (DOI: 10.21981/QADE-N936).

    BibTex | EndNote

Tags

  1. computational materials science
  2. linear regression
  3. machine learning
  4. neural networks
  5. physical science
  6. python
  7. supervised learning
  8. unsupervised learning