Deep regression learning with optimal loss function

TL;DR

This paper introduces a novel neural network-based nonparametric regression estimator that uses an optimal loss function derived from estimated likelihood, offering robustness, flexibility, and improved prediction accuracy over existing methods.

Contribution

The paper proposes a new adaptive loss function for neural network regression that is robust, distribution-agnostic, and theoretically optimal, with proven large sample properties.

Findings

Outperforms existing methods in prediction accuracy, efficiency, and robustness.

Comparable to the true MLE and improves with larger sample sizes.

Significantly reduces out-of-sample prediction errors in real data applications.

Abstract

In this paper, we develop a novel efficient and robust nonparametric regression estimator under a framework of feedforward neural network. There are several interesting characteristics for the proposed estimator. First, the loss function is built upon an estimated maximum likelihood function, who integrates the information from observed data, as well as the information from data structure. Consequently, the resulting estimator has desirable optimal properties, such as efficiency. Second, different from the traditional maximum likelihood estimation (MLE), the proposed method avoid the specification of the distribution, hence is flexible to any kind of distribution, such as heavy tails, multimodal or heterogeneous distribution. Third, the proposed loss function relies on probabilities rather than direct observations as in least squares, contributing the robustness in the proposed estimator. Finally, the proposed loss function involves nonparametric regression function only. This enables a direct application of existing packages, simplifying the computation and programming. We establish the large sample property of the proposed estimator in terms of its excess risk and minimax near-optimal rate. The theoretical results demonstrate that the proposed estimator is equivalent to the true MLE in which the density function is known. Our simulation studies show that the proposed estimator outperforms the existing methods in terms of prediction accuracy, efficiency and robustness. Particularly, it is comparable to the true MLE, and even gets better as the sample size increases. This implies that the adaptive and data-driven loss function from the estimated density may offer an additional avenue for capturing valuable information. We further apply the proposed method to four real data examples, resulting in significantly reduced out-of-sample prediction errors compared to existing methods.