@inproceedings{10.1145/3411764.3445538,
author = {Schoop, Eldon and Huang, Forrest and Hartmann, Bjoern},
title = {UMLAUT: Debugging Deep Learning Programs Using Program Structure and Model Behavior},
year = {2021},
isbn = {9781450380966},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
url = {https://doi.org/10.1145/3411764.3445538},
doi = {10.1145/3411764.3445538},
abstract = {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, andIn this work, we categorize the types of errors novices run into when writing ML code, and map them onto opportunities where tools could help. We 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 and were able to implement fixes for more bugs compared to a baseline condition. },
booktitle = {Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems},
articleno = {310},
numpages = {16},
keywords = {ML Debugging, ML Development, End-User ML},
location = {Yokohama, Japan},
series = {CHI '21}
}