Quantitative Overfitting Management for Human-in-the-loop ML Application Development with ease.ml/meter

TL;DR

This paper introduces ease.ml/meter, a system that automatically detects and measures overfitting during ML development, providing probabilistic guarantees and guiding data size decisions to improve model quality management.

Contribution

The paper presents a novel system for automated overfitting detection with probabilistic guarantees, addressing challenges in continuous ML development and adaptive analysis.

Findings

effectively detects overfitting with probabilistic guarantees

guides data size decisions for validation and testing

demonstrates practical tractability and effectiveness

Abstract

Simplifying machine learning (ML) application development, including distributed computation, programming interface, resource management, model selection, etc, has attracted intensive interests recently. These research efforts have significantly improved the efficiency and the degree of automation of developing ML models. In this paper, we take a first step in an orthogonal direction towards automated quality management for human-in-the-loop ML application development. We build ease. ml/meter, a system that can automatically detect and measure the degree of overfitting during the whole lifecycle of ML application development. ease. ml/meter returns overfitting signals with strong probabilistic guarantees, based on which developers can take appropriate actions. In particular, ease. ml/meter provides principled guidelines to simple yet nontrivial questions regarding desired validation and test data sizes, which are among commonest questions raised by developers. The fact that ML application development is typically a continuous procedure further worsens the situation: The validation and test data sets can lose their statistical power quickly due to multiple accesses, especially in the presence of adaptive analysis. ease. ml/meter addresses these challenges by leveraging a collection of novel techniques and optimizations, resulting in practically tractable data sizes without compromising the probabilistic guarantees. We present the design and implementation details of ease. ml/meter, as well as detailed theoretical analysis and empirical evaluation of its effectiveness.