Exploring the Relative Value of Collaborative Optimisation Pathways (Student Abstract)

Authors

  • Sudarshan Sreeram Imperial College London, London, United Kingdom

DOI:

https://doi.org/10.1609/aaai.v37i13.27028

Keywords:

Machine Learning, Model Compression, Keyword Spotting, Collaborative Optimisation, On-device Inference

Abstract

Compression techniques in machine learning (ML) independently improve a model’s inference efficiency by reducing its memory footprint while aiming to maintain its quality. This paper lays groundwork in questioning the merit of a compression pipeline involving all techniques as opposed to skipping a few by considering a case study on a keyword spotting model: DS-CNN-S. In addition, it documents improvements to the model’s training and dataset infrastructure. For this model, preliminary findings suggest that a full-scale pipeline isn’t required to achieve a competent memory footprint and accuracy, but a more comprehensive study is required.

Downloads

Published

2024-07-15

How to Cite

Sreeram, S. (2024). Exploring the Relative Value of Collaborative Optimisation Pathways (Student Abstract). Proceedings of the AAAI Conference on Artificial Intelligence, 37(13), 16336-16337. https://doi.org/10.1609/aaai.v37i13.27028