Two-Dimensional Deep ReLU CNN Approximation for Korobov Functions: A Constructive Approach
It provides a theoretical foundation for 2D CNNs in function approximation, which is incremental but important for broader applications in machine learning.
This paper tackles the problem of approximating Korobov functions using two-dimensional deep convolutional neural networks (CNNs) with a constructive approach, achieving near-optimal approximation rates that significantly reduce the curse of dimensionality.
This paper investigates approximation capabilities of two-dimensional (2D) deep convolutional neural networks (CNNs), with Korobov functions serving as a benchmark. We focus on 2D CNNs, comprising multi-channel convolutional layers with zero-padding and ReLU activations, followed by a fully connected layer. We propose a fully constructive approach for building 2D CNNs to approximate Korobov functions and provide rigorous analysis of the complexity of the constructed networks. Our results demonstrate that 2D CNNs achieve near-optimal approximation rates under the continuous weight selection model, significantly alleviating the curse of dimensionality. This work provides a solid theoretical foundation for 2D CNNs and illustrates their potential for broader applications in function approximation.