Unsupervised machine learning for physical concepts
This work addresses the challenge of enabling machines to learn scientific concepts from data, which could assist scientists in research, though it appears incremental as it builds on existing methods like Betti numbers and VAEs.
The paper tackles the problem of extracting interpretable physical concepts from experimental data using unsupervised machine learning, proposing a hybrid method that first computes Betti numbers and then uses a variational autoencoder to extract meaningful physical variables, demonstrating its functionality on toy models.
In recent years, machine learning methods have been used to assist scientists in scientific research. Human scientific theories are based on a series of concepts. How machine learns the concepts from experimental data will be an important first step. We propose a hybrid method to extract interpretable physical concepts through unsupervised machine learning. This method consists of two stages. At first, we need to find the Betti numbers of experimental data. Secondly, given the Betti numbers, we use a variational autoencoder network to extract meaningful physical variables. We test our protocol on toy models and show how it works.