Progressive Monitoring of Generative Model Training Evolution

TL;DR

This paper presents a progressive monitoring framework for deep generative models that uses dimensionality reduction to analyze training dynamics, enabling early detection of biases and failures to optimize training and improve output quality.

Contribution

The paper introduces a novel progressive analysis method for monitoring DGM training, facilitating early issue detection and bias mitigation through latent space and distribution analysis.

Findings

Effective early detection of biases during GAN training

Improved quality of generated data through timely interventions

Reduced computational resources by monitoring training progress

Abstract

While deep generative models (DGMs) have gained popularity, their susceptibility to biases and other inefficiencies that lead to undesirable outcomes remains an issue. With their growing complexity, there is a critical need for early detection of issues to achieve desired results and optimize resources. Hence, we introduce a progressive analysis framework to monitor the training process of DGMs. Our method utilizes dimensionality reduction techniques to facilitate the inspection of latent representations, the generated and real distributions, and their evolution across training iterations. This monitoring allows us to pause and fix the training method if the representations or distributions progress undesirably. This approach allows for the analysis of a models' training dynamics and the timely identification of biases and failures, minimizing computational loads. We demonstrate how our method supports identifying and mitigating biases early in training a Generative Adversarial Network (GAN) and improving the quality of the generated data distribution.