Resonance: Replacing Software Constants with Context-Aware Models in Real-time Communication
Abstract
Large software systems tune hundreds of 'constants' to optimize their runtime performance. These values are commonly derived through intuition, lab tests, or A/B tests. A 'one-size-fits-all' approach is often sub-optimal as the best value depends on runtime context. In this paper, we provide an experimental approach to replace constants with learned contextual functions for Skype - a widely used real-time communication (RTC) application. We present Resonance, a system based on contextual bandits (CB). We describe experiences from three real-world experiments: applying it to the audio, video, and transport components in Skype. We surface a unique and practical challenge of performing machine learning (ML) inference in large software systems written using encapsulation principles. Finally, we open-source FeatureBroker, a library to reduce the friction in adopting ML models in such development environments
Cite
@article{arxiv.2011.12715,
title = {Resonance: Replacing Software Constants with Context-Aware Models in Real-time Communication},
author = {Jayant Gupchup and Ashkan Aazami and Yaran Fan and Senja Filipi and Tom Finley and Scott Inglis and Marcus Asteborg and Luke Caroll and Rajan Chari and Markus Cozowicz and Vishak Gopal and Vinod Prakash and Sasikanth Bendapudi and Jack Gerrits and Eric Lau and Huazhou Liu and Marco Rossi and Dima Slobodianyk and Dmitri Birjukov and Matty Cooper and Nilesh Javar and Dmitriy Perednya and Sriram Srinivasan and John Langford and Ross Cutler and Johannes Gehrke},
journal= {arXiv preprint arXiv:2011.12715},
year = {2020}
}
Comments
Workshop on ML for Systems at NeurIPS 2020, Accepted