Thermodynamics of learning physical phenomena
It addresses the challenge of enhancing accuracy and credibility in ML for physics applications, but is incremental as it reviews existing insights.
The paper reviews how thermodynamics serves as an inductive bias to improve machine learning predictions for physical phenomena, exploring factors like scale, variable selection, and learning techniques.
Thermodynamics could be seen as an expression of physics at a high epistemic level. As such, its potential as an inductive bias to help machine learning procedures attain accurate and credible predictions has been recently realized in many fields. We review how thermodynamics provides helpful insights in the learning process. At the same time, we study the influence of aspects such as the scale at which a given phenomenon is to be described, the choice of relevant variables for this description or the different techniques available for the learning process.