Branes with brains: exploring string vacua with deep reinforcement learning

Abstract

We propose deep reinforcement learning as a model-free method for exploring the landscape of string vacua. As a concrete application, we utilize an artificial intelligence agent known as an asynchronous advantage actor-critic to explore type IIA compactifications with intersecting D6-branes. As different string background configurations are explored by changing D6-brane configurations, the agent receives rewards and punishments related to string consistency conditions and proximity to Standard Model vacua. These are in turn utilized to update the agent's policy and value neural networks to improve its behavior. By reinforcement learning, the agent's performance in both tasks is significantly improved, and for some tasks it finds a factor of O(200) more solutions than a random walker. In one case, we demonstrate that the agent learns a human-derived strategy for finding consistent string models. In another case, where no human-derived strategy exists, the agent learns a genuinely new strategy that achieves the same goal twice as efficiently per unit time. Our results demonstrate that the agent learns to solve various string theory consistency conditions simultaneously, which are phrased in terms of non-linear, coupled Diophantine equations.

Machine Learning Line Bundle Cohomology

“…Techniques of machine learning have recently been introduced into string theory and have been applied to a range of problems. Broadly, these applications can be divided into two classes, firstly, those which are attempting to facilitate difficult mathematical calculations required within string theory (“replacing the Mathematician”) and, secondly, those which employ machine learning techniques to deal with the vast amount of data string theory provides (“replacing the string theorist”).…”

Machine learning and serving of discrete field theories

“…As a specific example, methods of learning the Hamiltonian function of a canonical symplectic Hamiltonian system were proposed very recently [34][35][36][37][38][39][40][41] . (iv) Using neural networks to generate sampling data in statistical ensembles for calculating equilibrium properties of physical systems [42][43][44][45] .…”

Searching the landscape of flux vacua with genetic algorithms

“…The discrete landscape can be viewed as the complement of a continuum of seemingly consistent low-energy effective field theories (EFTs) that cannot descend from a string compactification, deemed the swampland [22,23]. While the latter has received much attention in recent years, progress in data science might allow for systematic studies of [24] and machine learning [25][26][27][28][29][30][31][32][33][34].…”