# Automating Staged Rollout with Reinforcement Learning

arXiv: https://arxiv.org/abs/2204.02189 · HTML: https://mlsanity.com/paper/2204.02189/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 2 of its 18 references that are in the map.
- First posted on arXiv 1 Apr 2022.

## Builds on (most cited first, 2 of 2)

- [Towards Continual Reinforcement Learning: A Review and Perspectives](https://mlsanity.com/paper/2012.13490/) (2020)
- [Performance-oriented DevOps: A Research Agenda](https://mlsanity.com/paper/1508.04752/) (2015)

## Built on by (most cited first, 1 of 1)

- [Governed Capability Evolution: Lifecycle-Time Compatibility Checking and Rollback for AI-Component-Based Systems, with a Proof-of-Concept Evaluation on Embodied Agents](https://mlsanity.com/paper/2604.08059/) (2026)

## Abstract (arXiv)

Staged rollout is a strategy of incrementally releasing software updates to portions of the user population in order to accelerate defect discovery without incurring catastrophic outcomes such as system wide outages. Some past studies have examined how to quantify and automate staged rollout, but stop short of simultaneously considering multiple product or process metrics explicitly. This paper demonstrates the potential to automate staged rollout with multi-objective reinforcement learning in order to dynamically balance stakeholder needs such as time to deliver new features and downtime incurred by failures due to latent defects.