Angler: Helping Machine Translation Practitioners Prioritize Model Improvements
Top Cited Papers
Open Access
- 19 April 2023
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 1-20
Abstract
Machine learning (ML) models can fail in unexpected ways in the real world, but not all model failures are equal. With finite time and resources, ML practitioners are forced to prioritize their model debugging and improvement efforts. Through interviews with 13 ML practitioners at Apple, we found that practitioners construct small targeted test sets to estimate an error’s nature, scope, and impact on users. We built on this insight in a case study with machine translation models, and developed Angler, an interactive visual analytics tool to help practitioners prioritize model improvements. In a user study with 7 machine translation experts, we used Angler to understand prioritization practices when the input space is infinite, and obtaining reliable signals of model quality is expensive. Our study revealed that participants could form more interesting and user-focused hypotheses for prioritization by analyzing quantitative summary statistics and qualitatively assessing data by reading sentences.Keywords
All Related Versions
This publication has 63 references indexed in Scilit:
- Visual Diagnosis of Tree Boosting MethodsIEEE Transactions on Visualization and Computer Graphics, 2017
- "Why Should I Trust You?"2016
- A Review on Evaluation Metrics for Data Classification EvaluationsInternational Journal of Data Mining & Knowledge Management Process, 2015
- Multivariate Density Estimation2015
- Quality management on Amazon Mechanical Turk2010
- Investigating statistical machine learning as a tool for software development2008
- Evaluating Evaluation Methods for Generation in the Presence of Variation2005
- NLTK2004
- Multidimensional binary search trees used for associative searchingCommunications of the ACM, 1975
- LIII. On lines and planes of closest fit to systems of points in spaceJournal of Computers in Education, 1901