Angler: Helping Machine Translation Practitioners Prioritize Model Improvements
This work addresses the challenge for machine translation practitioners in prioritizing debugging efforts when resources are limited, representing an incremental improvement in tool design.
The paper tackled the problem of helping machine translation practitioners prioritize model improvements by developing Angler, an interactive visual analytics tool, which enabled participants to form more interesting and user-focused hypotheses for prioritization.
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.