Module 5: Neural Networks for Regression and Classification
Category
Published on
Abstract
This module introduces neural networks for material science and engineering with hands-on online simulations. Neural networks are a subset of machine learning models used to learn mappings between inputs and outputs for a given dataset. Neural networks offer great flexibility and have shown great promise in the context of predicting material properties. In this module, we will explore neural networks to perform regression and classification tasks and predict the Young’s modulus and the crystal structure of some materials.
This end-to-end module is designed to be self-contained and easy to incorporate in existing courses or used for self-study. The module consists of three components:
- Pre-recoded lecture: introduction to neural networks in science and engineering
YouTube | Video Download (MP4) | Slides (PDF) - Hands-on tutorial using nanoHUB:
- Neural networks for regression
Download (PDF) - Neural networks for classification
Download (PDF)
- Neural networks for regression
- Homework Assignments:
- Regression
Download (PDF) - Classification
Download (PDF)
- Regression
This module is part of a series on data science and machine learning for engineering and physical sciences. Users will be able to run interactive code online using nanoHUB, no need to download or install any software.
Learning objectives. After completing this module, you will:
- Be able to create and train a neural network
- Be able to define objective functions for regression and classification tasks
- Know how to determine overfitting and underfitting in training neural networks
Pre-requisites
- Basic Python programming (see https://nanohub.org/resources/33266)
- Querying materials repositories (optional, see material in this series)
- Linear regression (optional, see material in this series)
Sponsored by
Cite this work
Researchers should cite this work as follows: