PSI Draft Specification

TL;DR

This paper introduces a draft specification for delivering machine learning services over HTTP, detailing essential components and motivations for ML as a service, as part of a broader inference protocols project.

Contribution

It provides a standardized draft specification for ML service delivery over HTTP, advancing the infrastructure for machine learning as a service.

Findings

Defines essential components for ML service protocols

Outlines optional features for flexible ML service deployment

Supports interoperability in machine learning service implementations

Abstract

This document presents the draft specification for delivering machine learning services over HTTP, developed as part of the Protocols and Structures for Inference project, which concluded in 2013. It presents the motivation for providing machine learning as a service, followed by a description of the essential and optional components of such a service.