Which Hyperparameters Matter? A Game-Theoretic Framework for Interpretable Hyperparameter Sensitivity Analysis
Machine Learning
Summary
The authors present a way to understand how different hyperparameters affect multiple goals in machine learning models using ideas from game theory. They use Shapley Effects to measure how important each hyperparameter is overall, and Pareto front sets to find the best hyperparameter combinations early on. This helps users see which settings matter most for their specific goals, making it easier to choose good hyperparameters without testing everything. They show that their method works well with several kinds of neural networks and tasks involving multiple objectives.
Authors
Nyi Nyi Aung, Heepeom Shin, Abigail Lawlor, Adrian Stein
Abstract
This work presents a game-theoretic framework for interpretable hyperparameter-objective interaction analysis rather than proposing a new optimization algorithm. In the proposed framework, Shapley Effects are employed for global sensitivity analysis, while Pareto front sets are utilized to identify effective hyperparameter configurations and support early-stage model evaluation. The resulting analysis reveals which players (hyperparameters) are most influential with respect to different objectives in a given game (application). Consequently, the proposed framework provides interpretable insights into objective-aware hyperparameter interactions, enabling practitioners to guide subsequent optimization, reduce the search space, and perform early-stage model evaluation. The effectiveness of the proposed framework is demonstrated using three distinct neural network architectures across different problem domains under multi-objective settings.