Improved Bounds on Neural Complexity for Representing Piecewise Linear Functions

TL;DR

This paper establishes tighter bounds on the neural complexity needed to represent piecewise linear functions, showing polynomial bounds invariant to input dimension and providing an efficient algorithm for such representations.

Contribution

It introduces new polynomial bounds on the number of neurons for CPWL functions, improving upon previous exponential estimates, and offers a polynomial-time algorithm for constructing such networks.

Findings

Neural complexity is at most quadratic in the number of pieces.

Complexity becomes bilinear when considering the number of linear components.

Bounds are invariant to input dimension, unlike previous results.

Abstract

A deep neural network using rectified linear units represents a continuous piecewise linear (CPWL) function and vice versa. Recent results in the literature estimated that the number of neurons needed to exactly represent any CPWL function grows exponentially with the number of pieces or exponentially in terms of the factorial of the number of distinct linear components. Moreover, such growth is amplified linearly with the input dimension. These existing results seem to indicate that the cost of representing a CPWL function is expensive. In this paper, we propose much tighter bounds and establish a polynomial time algorithm to find a network satisfying these bounds for any given CPWL function. We prove that the number of hidden neurons required to exactly represent any CPWL function is at most a quadratic function of the number of pieces. In contrast to all previous results, this upper bound is invariant to the input dimension. Besides the number of pieces, we also study the number of distinct linear components in CPWL functions. When such a number is also given, we prove that the quadratic complexity turns into bilinear, which implies a lower neural complexity because the number of distinct linear components is always not greater than the minimum number of pieces in a CPWL function. When the number of pieces is unknown, we prove that, in terms of the number of distinct linear components, the neural complexities of any CPWL function are at most polynomial growth for low-dimensional inputs and factorial growth for the worst-case scenario, which are significantly better than existing results in the literature.