The Hintons in your Neural Network: a Quantum Field Theory View of Deep Learning

In this work we develop a quantum field theory formalism for deep learning,\nwhere input signals are encoded in Gaussian states, a generalization of\nGaussian processes which encode the agent's uncertainty about the input signal.\nWe show how to represent linear and non-linear layers as unitary quantum gates,\nand interpret the fundamental excitations of the quantum model as particles,\ndubbed ``Hintons''. On top of opening a new perspective and techniques for\nstudying neural networks, the quantum formulation is well suited for optical\nquantum computing, and provides quantum deformations of neural networks that\ncan be run efficiently on those devices. Finally, we discuss a semi-classical\nlimit of the quantum deformed models which is amenable to classical simulation.\n