Git-Theta: A Git Extension for Collaborative Development of Machine Learning Models

Nikhil Kandpal, Brian Lester, Mohammed Muqeeth, Anisha Mascarenhas, Monty Evans, Vishal Baskaran, Tenghao Huang, Haokun Liu, Colin Raffel
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:15708-15719, 2023.

Abstract

Currently, most machine learning models are trained by centralized teams and are rarely updated. In contrast, open-source software development involves the iterative development of a shared artifact through distributed collaboration using a version control system. In the interest of enabling collaborative and continual improvement of machine learning models (Raffel, 2023), we introduce Git-Theta, a version control system for machine learning models. Git-Theta is an extension to Git, the most widely used version control software, that allows fine-grained tracking of changes to model parameters alongside code and other artifacts. Unlike existing version control systems that treat a model checkpoint as a blob of data, Git-Theta leverages the structure of checkpoints to support communication-efficient updates, automatic model merges, and meaningful reporting about the difference between two versions of a model. In addition, Git-Theta includes a plug-in system that enables users to easily add support for new functionality. In this paper, we introduce Git-Theta’s design and features and include an example use-case of Git-Theta where a pre-trained model is continually adapted and modified. We publicly release Git-Theta in hopes of kickstarting a new era of collaborative model development. https://github.com/r-three/git-theta/

Cite this Paper

BibTeX
@InProceedings{pmlr-v202-kandpal23b, title = {Git-Theta: A Git Extension for Collaborative Development of Machine Learning Models}, author = {Kandpal, Nikhil and Lester, Brian and Muqeeth, Mohammed and Mascarenhas, Anisha and Evans, Monty and Baskaran, Vishal and Huang, Tenghao and Liu, Haokun and Raffel, Colin}, booktitle = {Proceedings of the 40th International Conference on Machine Learning}, pages = {15708--15719}, year = {2023}, editor = {Krause, Andreas and Brunskill, Emma and Cho, Kyunghyun and Engelhardt, Barbara and Sabato, Sivan and Scarlett, Jonathan}, volume = {202}, series = {Proceedings of Machine Learning Research}, month = {23--29 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v202/kandpal23b/kandpal23b.pdf}, url = {https://proceedings.mlr.press/v202/kandpal23b.html}, abstract = {Currently, most machine learning models are trained by centralized teams and are rarely updated. In contrast, open-source software development involves the iterative development of a shared artifact through distributed collaboration using a version control system. In the interest of enabling collaborative and continual improvement of machine learning models (Raffel, 2023), we introduce Git-Theta, a version control system for machine learning models. Git-Theta is an extension to Git, the most widely used version control software, that allows fine-grained tracking of changes to model parameters alongside code and other artifacts. Unlike existing version control systems that treat a model checkpoint as a blob of data, Git-Theta leverages the structure of checkpoints to support communication-efficient updates, automatic model merges, and meaningful reporting about the difference between two versions of a model. In addition, Git-Theta includes a plug-in system that enables users to easily add support for new functionality. In this paper, we introduce Git-Theta’s design and features and include an example use-case of Git-Theta where a pre-trained model is continually adapted and modified. We publicly release Git-Theta in hopes of kickstarting a new era of collaborative model development. https://github.com/r-three/git-theta/} }
Endnote
%0 Conference Paper %T Git-Theta: A Git Extension for Collaborative Development of Machine Learning Models %A Nikhil Kandpal %A Brian Lester %A Mohammed Muqeeth %A Anisha Mascarenhas %A Monty Evans %A Vishal Baskaran %A Tenghao Huang %A Haokun Liu %A Colin Raffel %B Proceedings of the 40th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2023 %E Andreas Krause %E Emma Brunskill %E Kyunghyun Cho %E Barbara Engelhardt %E Sivan Sabato %E Jonathan Scarlett %F pmlr-v202-kandpal23b %I PMLR %P 15708--15719 %U https://proceedings.mlr.press/v202/kandpal23b.html %V 202 %X Currently, most machine learning models are trained by centralized teams and are rarely updated. In contrast, open-source software development involves the iterative development of a shared artifact through distributed collaboration using a version control system. In the interest of enabling collaborative and continual improvement of machine learning models (Raffel, 2023), we introduce Git-Theta, a version control system for machine learning models. Git-Theta is an extension to Git, the most widely used version control software, that allows fine-grained tracking of changes to model parameters alongside code and other artifacts. Unlike existing version control systems that treat a model checkpoint as a blob of data, Git-Theta leverages the structure of checkpoints to support communication-efficient updates, automatic model merges, and meaningful reporting about the difference between two versions of a model. In addition, Git-Theta includes a plug-in system that enables users to easily add support for new functionality. In this paper, we introduce Git-Theta’s design and features and include an example use-case of Git-Theta where a pre-trained model is continually adapted and modified. We publicly release Git-Theta in hopes of kickstarting a new era of collaborative model development. https://github.com/r-three/git-theta/
APA
Kandpal, N., Lester, B., Muqeeth, M., Mascarenhas, A., Evans, M., Baskaran, V., Huang, T., Liu, H. & Raffel, C.. (2023). Git-Theta: A Git Extension for Collaborative Development of Machine Learning Models. Proceedings of the 40th International Conference on Machine Learning, in Proceedings of Machine Learning Research 202:15708-15719 Available from https://proceedings.mlr.press/v202/kandpal23b.html.

Related Material