Data Augmentation for Deep Learning Regression Tasks by Machine Learning Models

2025-01-07

AI Summary

Deep learning models often exhibit weak generalization and subpar performance on tabular regression tasks compared to traditional methods. To address this, we propose a novel data augmentation framework that preserves the original statistical relationships in tabular data. Our approach integrates controllable noise injection, statistics-driven resampling, and distributional consistency constraints—outperforming baseline strategies such as simple duplication or Gaussian noise addition. We conduct a systematic evaluation across 30 diverse real-world regression datasets, leveraging automated deep learning (AutoDL) frameworks—including AutoKeras, H2O, and AutoGluon—for end-to-end training and validation. Results demonstrate an average improvement of over 10% in neural network regression performance, enabling enhanced models to consistently surpass their un-augmented counterparts. The method exhibits strong generalizability and robustness across heterogeneous domains. Importantly, it provides a scalable, reproducible pathway for deploying deep learning in practical tabular regression applications.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Natural Language Processing: Generation
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Deep learning (DL) models have gained prominence in domains such as computer vision and natural language processing but remain underutilized for regression tasks involving tabular data. In these cases, traditional machine learning (ML) models often outperform DL models. In this study, we propose and evaluate various data augmentation (DA) techniques to improve the performance of DL models for tabular data regression tasks. We compare the performance gain of Neural Networks by different DA strategies ranging from a naive method of duplicating existing observations and adding noise to a more sophisticated DA strategy that preserves the underlying statistical relationship in the data. Our analysis demonstrates that the advanced DA method significantly improves DL model performance across multiple datasets and regression tasks, resulting in an average performance increase of over 10% compared to baseline models without augmentation. The efficacy of these DA strategies was rigorously validated across 30 distinct datasets, with multiple iterations and evaluations using three different automated deep learning (AutoDL) frameworks: AutoKeras, H2O, and AutoGluon. This study demonstrates that by leveraging advanced DA techniques, DL models can realize their full potential in regression tasks, thereby contributing to broader adoption and enhanced performance in practical applications.

Problem

• Deep Learning
• Tabular Data
• Prediction Accuracy

Innovation

• Complex Data Augmentation
• Deep Learning Model
• Tabular Data Prediction