Entropy from Machine Learning
This provides a new computational tool for physicists and researchers analyzing entropy in systems like spin models or neural data, though it is incremental as it adapts existing ML methods to a specific problem.
The authors tackled the problem of calculating entropy for binary configurations by translating it into supervised classification tasks, enabling the use of any machine learning classifier to compute entropy and free energy from Monte Carlo data; they successfully reproduced the entropy and free energy of the 2D Ising model across temperatures.
We translate the problem of calculating the entropy of a set of binary configurations/signals into a sequence of supervised classification tasks. Subsequently, one can use virtually any machine learning classification algorithm for computing entropy. This procedure can be used to compute entropy, and consequently the free energy directly from a set of Monte Carlo configurations at a given temperature. As a test of the proposed method, using an off-the-shelf machine learning classifier we reproduce the entropy and free energy of the 2D Ising model from Monte Carlo configurations at various temperatures throughout its phase diagram. Other potential applications include computing the entropy of spiking neurons or any other multidimensional binary signals.