Fixed point actions from convolutional neural networks
This work addresses lattice gauge theory simulations for physicists, offering a potentially incremental improvement in accuracy for fixed point actions.
The paper tackled the problem of lattice artifacts in lattice gauge theory by using lattice gauge-equivariant convolutional neural networks (L-CNNs) to parametrize fixed point actions, resulting in much higher accuracy compared to older methods and enabling scale-invariant instanton solutions with minimal artifacts on coarse lattices.
Lattice gauge-equivariant convolutional neural networks (L-CNNs) can be used to form arbitrarily shaped Wilson loops and can approximate any gauge-covariant or gauge-invariant function on the lattice. Here we use L-CNNs to describe fixed point (FP) actions which are based on renormalization group transformations. FP actions are classically perfect, i.e., they have no lattice artifacts on classical gauge-field configurations satisfying the equations of motion, and therefore possess scale invariant instanton solutions. FP actions are tree-level Symanzik-improved to all orders in the lattice spacing and can produce physical predictions with very small lattice artifacts even on coarse lattices. We find that L-CNNs are much more accurate at parametrizing the FP action compared to older approaches. They may therefore provide a way to circumvent critical slowing down and topological freezing towards the continuum limit.