UMLAUT: Debugging Deep Learning Programs using Program Structure and Model Behavior

要旨

Training deep neural networks can generate non-descriptive error messages or produce unusual output without any explicit errors at all. While experts rely on tacit knowledge to apply debugging strategies, non-experts lack the experience required to interpret model output and correct Deep Learning (DL) programs. In this work, we identify DL debugging heuristics and strategies used by experts, and use them to guide the design of Umlaut. Umlaut checks DL program structure and model behavior against these heuristics; provides human-readable error messages to users; and annotates erroneous model output to facilitate error correction. Umlaut links code, model output, and tutorial-driven error messages in a single interface. We evaluated Umlaut in a study with 15 participants to determine its effectiveness in helping developers find and fix errors in their DL programs. Participants using Umlaut found and fixed significantly more bugs compared to a baseline condition.

著者
Eldon Schoop
University of California, Berkeley, Berkeley, California, United States
Forrest Huang
University of California, Berkeley, Berkeley, California, United States
Bjoern Hartmann
UC Berkeley, Berkeley, California, United States
DOI

10.1145/3411764.3445538

論文URL

https://doi.org/10.1145/3411764.3445538

動画

会議: CHI 2021

The ACM CHI Conference on Human Factors in Computing Systems (https://chi2021.acm.org/)

セッション: Engineering Development Support

[A] Paper Room 05, 2021-05-10 17:00:00~2021-05-10 19:00:00 / [B] Paper Room 05, 2021-05-11 01:00:00~2021-05-11 03:00:00 / [C] Paper Room 05, 2021-05-11 09:00:00~2021-05-11 11:00:00
Paper Room 05
14 件の発表
2021-05-10 17:00:00
2021-05-10 19:00:00
日本語まとめ
読み込み中…