File(s) under permanent embargo
Protocols and structures for inference: a RESTful API for machine learning
conference contribution
posted on 2023-05-23, 10:09 authored by Erin MontgomeryErin Montgomery, Reid, MD, Drake, BDiversity in machine learning APIs (in both software toolkits and web services), works against realising machine learning’s full potential, making it difficult to draw on individual algorithms from different products or to compose multiple algorithms to solve complex tasks. This paper introduces the Protocols and Structures for Inference (PSI) service architecture and specification, which presents inferential entities - relations, attributes, learners and predictors - as RESTful web resources that are accessible via a common but flexible and extensible interface. Resources describe the data they ingest or emit using a variant of the JSON schema language, and the API has mechanisms to support non-JSON data and future extension of service features.
History
Publication title
Proceedings of Machine Learning Research (PMLR) - Volume 50: 2nd Conference on Predictive APIs and Apps (PAPIs '15)Volume
50Editors
L Dorard, MD Reid & FJ MartinPagination
29-42ISSN
1532-4435Department/School
School of Information and Communication TechnologyPublisher
Microtome PublishingPlace of publication
Brookline, MA USAEvent title
2nd International Conference on Predictive APIs and Apps (PAPIs '15)Event Venue
Sydney, AustraliaDate of Event (Start Date)
2015-08-06Date of Event (End Date)
2015-08-07Rights statement
Copyright 2016 The authorsRepository Status
- Restricted