Summary of Establishing Deep Infomax As An Effective Self-supervised Learning Methodology in Materials Informatics, by Michael Moran et al.
Establishing Deep InfoMax as an effective self-supervised learning methodology in materials informatics
by Michael Moran, Vladimir V. Gusev, Michael W. Gaultois, Dmytro Antypov, Matthew J. Rosseinsky
First submitted to arxiv on: 30 Jun 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Materials Science (cond-mat.mtrl-sci)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed research aims to improve property prediction results in materials informatics by leveraging large amounts of crystal data without property labels. A self-supervised machine learning framework called Deep InfoMax is used to pretrain supervised models on these unlabeled datasets, maximizing the mutual information between a point set representation of a crystal and a vector representation suitable for downstream learning. The effectiveness of this approach is demonstrated in the contexts of representation learning and transfer learning for tasks such as band gap and formation energy prediction. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This research solves a big problem in materials informatics by showing how to use lots of data without property labels to improve predictions on smaller datasets. Scientists can now train models on huge amounts of crystal data, even if they don’t have labels for every material. This means that scientists can make better predictions about the properties of new materials with very little data. |
Keywords
» Artificial intelligence » Machine learning » Representation learning » Self supervised » Supervised » Transfer learning
