# How to "DODGE" Complex Software Analytics?

arXiv: https://arxiv.org/abs/1902.01838 · HTML: https://mlsanity.com/paper/1902.01838/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 5 Feb 2019.

## Built on by (most cited first, 11 of 11)

- [FRUGAL: Unlocking SSL for Software Analytics](https://mlsanity.com/paper/2108.09847/) (2021)
- [Learning from Very Little Data: On the Value of Landscape Analysis for Predicting Software Project Health](https://mlsanity.com/paper/2301.06577/) (2023)
- [Old but Gold: Reconsidering the value of feedforward learners for software analytics](https://mlsanity.com/paper/2101.06319/) (2021)
- [How to Find Actionable Static Analysis Warnings: A Case Study with FindBugs](https://mlsanity.com/paper/2205.10504/) (2022)
- [Is Hyper-Parameter Optimization Different for Software Analytics?](https://mlsanity.com/paper/2401.09622/) (2024)
- [Software Engineering Principles for Fairer Systems: Experiments with GroupCART](https://mlsanity.com/paper/2504.12587/) (2025)
- [Assessing the Early Bird Heuristic (for Predicting Project Quality)](https://mlsanity.com/paper/2105.11082/) (2021)
- [An Expert System for Redesigning Software for Cloud Applications](https://mlsanity.com/paper/2109.14569/) (2021)
- [Less, but Stronger: On the Value of Strong Heuristics in Semi-supervised Learning for Software Analytics](https://mlsanity.com/paper/2302.01997/) (2023)
- [Strong convexity-guided hyper-parameter optimization for flatter losses](https://mlsanity.com/paper/2402.05025/) (2024)
- [Is Model Instability just Noise to be Tolerated or a Property that can be Managed?](https://mlsanity.com/paper/2607.10420/) (2026)

## Abstract (arXiv)

Machine learning techniques applied to software engineering tasks can be improved by hyperparameter optimization, i.e., automatic tools that find good settings for a learner's control parameters. We show that such hyperparameter optimization can be unnecessarily slow, particularly when the optimizers waste time exploring "redundant tunings"', i.e., pairs of tunings which lead to indistinguishable results. By ignoring redundant tunings, DODGE, a tuning tool, runs orders of magnitude faster, while also generating learners with more accurate predictions than seen in prior state-of-the-art approaches.