# Machine Translation System Selection from Bandit Feedback

arXiv: https://arxiv.org/abs/2002.09646 · HTML: https://mlsanity.com/paper/2002.09646/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 22 Feb 2020.

## Built on by (most cited first, 4 of 4)

- [Bandits Don't Follow Rules: Balancing Multi-Facet Machine Translation with Multi-Armed Bandits](https://mlsanity.com/paper/2110.06997/) (2021)
- [Online Learning Meets Machine Translation Evaluation: Finding the Best Systems with the Least Human Effort](https://mlsanity.com/paper/2105.13385/) (2021)
- [Onception: Active Learning with Expert Advice for Real World Machine Translation](https://mlsanity.com/paper/2203.04507/) (2022)
- [EvolveMT: an Ensemble MT Engine Improving Itself with Usage Only](https://mlsanity.com/paper/2306.11823/) (2023)

## Abstract (arXiv)

Adapting machine translation systems in the real world is a difficult problem. In contrast to offline training, users cannot provide the type of fine-grained feedback (such as correct translations) typically used for improving the system. Moreover, different users have different translation needs, and even a single user's needs may change over time. In this work we take a different approach, treating the problem of adaptation as one of selection. Instead of adapting a single system, we train many translation systems using different architectures, datasets, and optimization methods. Using bandit learning techniques on simulated user feedback, we learn a policy to choose which system to use for a particular translation task. We show that our approach can (1) quickly adapt to address domain changes in translation tasks, (2) outperform the single best system in mixed-domain translation tasks, and (3) make effective instance-specific decisions when using contextual bandit strategies.