Artificial Neural Networks as Trial Wave Functions for Quantum Monte Carlo

Inspired by the universal approximation theorem and widespread adoption of artificial neural network techniques in a diversity fields, we propose feed-forward networks as general purpose trial wave function for quantum Monte Carlo simulations continous many-body systems. Whereas simple model systems whole can be represented network, antisymmetry condition non-trivial fermionic is incorporated means Slater determinant. To demonstrate accuracy our functions, have studied an exactly solvable system two trapped interacting particles, well hydrogen dimer.