Kernel Limit for a Class of Recurrent Neural Networks Trained on Ergodic Data Sequences

TL;DR

This paper develops mathematical methods to analyze the asymptotic behavior of recurrent neural networks (RNNs) with simplified weights, revealing their convergence to infinite-dimensional limits and establishing neural tangent kernel (NTK) limits for large-scale training.

Contribution

It introduces a fixed point analysis for RNN memory states and derives NTK limits for RNNs trained on ergodic data sequences, addressing challenges unique to RNNs.

Findings

RNNs converge to solutions of infinite-dimensional ODEs and fixed point equations.

Established NTK limits for RNNs with growing data and network size.

Provided convergence estimates for RNN memory state evolution.

Abstract

Mathematical methods are developed to characterize the asymptotics of recurrent neural networks (RNN) as the number of hidden units, data samples in the sequence, hidden state updates, and training steps simultaneously grow to infinity. In the case of an RNN with a simplified weight matrix, we prove the convergence of the RNN to the solution of an infinite-dimensional ODE coupled with the fixed point of a random algebraic equation. The analysis requires addressing several challenges which are unique to RNNs. In typical mean-field applications (e.g., feedforward neural networks), discrete updates are of magnitude $\mathcal{O}(1/N)$ and the number of updates is $\mathcal{O}(N)$. Therefore, the system can be represented as an Euler approximation of an appropriate ODE/PDE, which it will converge to as $N \rightarrow \infty$. However, the RNN hidden layer updates are $\mathcal{O}(1)$. Therefore, RNNs cannot be represented as a discretization of an ODE/PDE and standard mean-field techniques cannot be applied. Instead, we develop a fixed point analysis for the evolution of the RNN memory states, with convergence estimates in terms of the number of update steps and the number of hidden units. The RNN hidden layer is studied as a function in a Sobolev space, whose evolution is governed by the data sequence (a Markov chain), the parameter updates, and its dependence on the RNN hidden layer at the previous time step. Due to the strong correlation between updates, a Poisson equation must be used to bound the fluctuations of the RNN around its limit equation. These mathematical methods give rise to the neural tangent kernel (NTK) limits for RNNs trained on data sequences as the number of data samples and size of the neural network grow to infinity.