# Helix: Holistic Optimization for Accelerating Iterative Machine Learning

arXiv: https://arxiv.org/abs/1812.05762 · HTML: https://mlsanity.com/paper/1812.05762/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 14 Dec 2018.

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

- [Production Machine Learning Pipelines: Empirical Analysis and Optimization Opportunities](https://mlsanity.com/paper/2103.16007/) (2021)
- [Whither AutoML? Understanding the Role of Automation in Machine Learning Workflows](https://mlsanity.com/paper/2101.04834/) (2021)
- [OmniFair: A Declarative System for Model-Agnostic Group Fairness in Machine Learning](https://mlsanity.com/paper/2103.09055/) (2021)
- [Distributed Learning Systems with First-order Methods](https://mlsanity.com/paper/2104.05245/) (2021)
- [SPIO: Ensemble and Selective Strategies via LLM-Based Multi-Agent Planning in Automated Data Science](https://mlsanity.com/paper/2503.23314/) (2025)
- [AutoML in Heavily Constrained Applications](https://mlsanity.com/paper/2306.16913/) (2023)
- [Compliance Cards: Automated EU AI Act Compliance Analyses amidst a Complex AI Supply Chain](https://mlsanity.com/paper/2406.14758/) (2024)
- [DeepEverest: Accelerating Declarative Top-K Queries for Deep Neural Network Interpretation](https://mlsanity.com/paper/2104.02234/) (2021)
- [stratum: A System Infrastructure for Massive Agent-Centric ML Workloads](https://mlsanity.com/paper/2603.03589/) (2026)
- [AI Factories: It's time to rethink the Cloud-HPC divide](https://mlsanity.com/paper/2509.12849/) (2025)
- [Gypscie: A Cross-Platform AI Artifact Management System](https://mlsanity.com/paper/2604.10311/) (2026)

## Abstract (arXiv)

Machine learning workflow development is a process of trial-and-error: developers iterate on workflows by testing out small modifications until the desired accuracy is achieved. Unfortunately, existing machine learning systems focus narrowly on model training---a small fraction of the overall development time---and neglect to address iterative development. We propose Helix, a machine learning system that optimizes the execution across iterations---intelligently caching and reusing, or recomputing intermediates as appropriate. Helix captures a wide variety of application needs within its Scala DSL, with succinct syntax defining unified processes for data preprocessing, model specification, and learning. We demonstrate that the reuse problem can be cast as a Max-Flow problem, while the caching problem is NP-Hard. We develop effective lightweight heuristics for the latter. Empirical evaluation shows that Helix is not only able to handle a wide variety of use cases in one unified workflow but also much faster, providing run time reductions of up to 19x over state-of-the-art systems, such as DeepDive or KeystoneML, on four real-world applications in natural language processing, computer vision, social and natural sciences.