Models derived from other models are extremely common in machine learning (ML) today. For example, transfer learning is used to create task-specific models from “pre-trained” models through finetuning. This has led to an ecosystem where models are related to each other, sharing structure and often even parameter values. However, it is hard to manage these model derivatives: the storage overhead of storing all derived models quickly becomes onerous, prompting users to get rid of intermediate models that might be useful for further analysis. Additionally, undesired behaviors in models are hard to track down (e.g., is a bug inherited from an upstream model?). In this paper, we propose a model versioning and management system called MGit that makes it easier to store, test, update, and collaborate on model derivatives. MGit introduces a lineage graph that records provenance and versioning information between models, optimizations to efficiently store model parameters, as well as abstractions over this lineage graph that facilitate relevant testing, updating and collaboration functionality. MGit is able to reduce the lineage graph’s storage footprint by up to 7x and automatically update downstream models in response to updates to upstream models.<\/p>\n","protected":false},"excerpt":{"rendered":"
Models derived from other models are extremely common in machine learning (ML) today. For example, transfer learning is used to create task-specific models from “pre-trained” models through finetuning. This has led to an ecosystem where models are related to each other, sharing structure and often even parameter values. However, it is hard to manage these […]<\/p>\n","protected":false},"featured_media":0,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"_classifai_error":"","footnotes":""},"msr-content-type":[3],"msr-research-highlight":[],"research-area":[13556,13547],"msr-publication-type":[193718],"msr-product-type":[],"msr-focus-area":[],"msr-platform":[],"msr-download-source":[],"msr-locale":[268875],"msr-post-option":[],"msr-field-of-study":[],"msr-conference":[],"msr-journal":[],"msr-impact-theme":[],"msr-pillar":[],"class_list":["post-963933","msr-research-item","type-msr-research-item","status-publish","hentry","msr-research-area-artificial-intelligence","msr-research-area-systems-and-networking","msr-locale-en_us"],"msr_publishername":"","msr_edition":"","msr_affiliation":"","msr_published_date":"2023-7-14","msr_host":"","msr_duration":"","msr_version":"","msr_speaker":"","msr_other_contributors":"","msr_booktitle":"","msr_pages_string":"","msr_chapter":"","msr_isbn":"","msr_journal":"","msr_volume":"","msr_number":"arXiv:2307.07507","msr_editors":"","msr_series":"","msr_issue":"","msr_organization":"arXiv","msr_how_published":"","msr_notes":"","msr_highlight_text":"","msr_release_tracker_id":"","msr_original_fields_of_study":"","msr_download_urls":"","msr_external_url":"","msr_secondary_video_url":"","msr_longbiography":"","msr_microsoftintellectualproperty":1,"msr_main_download":"","msr_publicationurl":"","msr_doi":"","msr_publication_uploader":[{"type":"url","viewUrl":"false","id":"false","title":"https:\/\/arxiv.org\/abs\/2307.07507","label_id":"243109","label":0}],"msr_related_uploader":"","msr_attachments":[],"msr-author-ordering":[{"type":"text","value":"Wei Hao","user_id":0,"rest_url":false},{"type":"text","value":"Daniel Mendoza","user_id":0,"rest_url":false},{"type":"user_nicename","value":"Rafael da Silva","user_id":40579,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Rafael da Silva"},{"type":"text","value":"Deepak Narayanan","user_id":0,"rest_url":false},{"type":"user_nicename","value":"Amar Phanishayee","user_id":30975,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Amar Phanishayee"}],"msr_impact_theme":[],"msr_research_lab":[199565],"msr_event":[],"msr_group":[144927],"msr_project":[472845],"publication":[],"video":[],"download":[],"msr_publication_type":"techreport","related_content":{"projects":[{"ID":472845,"post_title":"Project Fiddle","post_name":"fiddle","post_type":"msr-project","post_date":"2018-03-20 10:20:05","post_modified":"2025-02-06 11:06:39","post_status":"publish","permalink":"https:\/\/www.microsoft.com\/en-us\/research\/project\/fiddle\/","post_excerpt":"The goal of Project Fiddle is to build efficient systems infrastructure for very fast distributed DNN training. Our goal is to support 100x more efficient training. Our innovations cut across the systems stack: the memory subsystem, structuring parallel computation across GPUs and machines, and interconnects between GPUs and across machines.","_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-project\/472845"}]}}]},"_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/963933","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-research-item"}],"version-history":[{"count":1,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/963933\/revisions"}],"predecessor-version":[{"id":963936,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/963933\/revisions\/963936"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=963933"}],"wp:term":[{"taxonomy":"msr-content-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-content-type?post=963933"},{"taxonomy":"msr-research-highlight","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-highlight?post=963933"},{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=963933"},{"taxonomy":"msr-publication-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-publication-type?post=963933"},{"taxonomy":"msr-product-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-product-type?post=963933"},{"taxonomy":"msr-focus-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-focus-area?post=963933"},{"taxonomy":"msr-platform","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-platform?post=963933"},{"taxonomy":"msr-download-source","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-download-source?post=963933"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=963933"},{"taxonomy":"msr-post-option","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-post-option?post=963933"},{"taxonomy":"msr-field-of-study","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-field-of-study?post=963933"},{"taxonomy":"msr-conference","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-conference?post=963933"},{"taxonomy":"msr-journal","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-journal?post=963933"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=963933"},{"taxonomy":"msr-pillar","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-pillar?post=963933"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}