On Representing Electronic Wave Functions with Sign Equivariant Neural Networks

Nicholas Gao,Stephan Günnemann

2024-03-08

Abstract:Recent neural networks demonstrated impressively accurate approximations of electronic ground-state wave functions. Such neural networks typically consist of a permutation-equivariant neural network followed by a permutation-antisymmetric operation to enforce the electronic exchange symmetry. While accurate, such neural networks are computationally expensive. In this work, we explore the flipped approach, where we first compute antisymmetric quantities based on the electronic coordinates and then apply sign equivariant neural networks to preserve the antisymmetry. While this approach promises acceleration thanks to the lower-dimensional representation, we demonstrate that it reduces to a Jastrow factor, a commonly used permutation-invariant multiplicative factor in the wave function. Our empirical results support this further, finding little to no improvements over baselines. We conclude with neither theoretical nor empirical advantages of sign equivariant functions for representing electronic wave functions within the evaluation of this work.

Quantum Physics,Machine Learning,Chemical Physics,Computational Physics

What problem does this paper attempt to address?

This paper discusses how to use neural networks to represent electronic wave functions more efficiently, which is an important problem in quantum chemistry. The traditional approach is to use symmetric networks with anti-symmetric operations to enforce electron exchange symmetry, but this method is computationally expensive. The paper proposes a new method, which first calculates the anti-symmetric quantity and then applies a sign-equivariant neural network to maintain anti-symmetry, with the expectation of reducing computational complexity.

However, theoretical analysis shows that this new method is actually equivalent to the classical Jastrow factor, which means it does not reduce the number of ground state solutions required to represent wave functions. Experimental evidence also supports this theoretical finding, showing no significant improvements or advantages compared to the baseline method. Therefore, the authors conclude that in the current evaluation, both theoretically and empirically, the sign-equivariant function does not have a clear advantage in representing electronic wave functions.

The paper also discusses related previous work, including methods using neural networks to improve orbital functions for better wave function approximation accuracy, as well as optimization problems and numerical stability challenges associated with them. The experimental section compares the impact of different methods on energy and points out that in certain cases, using the proposed odd function may lead to optimization difficulties and performance degradation.

Overall, this paper proposes a new method to represent electronic wave functions, but studies show that this method is not superior to existing techniques and may introduce additional complexity and instability factors.