TY - GEN
T1 - Optimizing Machine Learning Inference Queries for Multiple Objectives
AU - Li, Ziyu
AU - Schonfeld, Mariette
AU - Hai, Rihan
AU - Bozzon, Alessandro
AU - Katsifodimos, Asterios
N1 - Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.
PY - 2023
Y1 - 2023
N2 - Given a set of pre-trained Machine Learning (ML) models, can we solve complex analytic tasks that make use of those models by formulating ML inference queries? Can we mitigate different tradeoffs, e.g., high accuracy, low execution costs and memory footprint, when optimizing the queries? In this work we present different multi-objective ML inference query optimization strategies, and compare them on their usability, applicability, and complexity. We formulate Mixed-Integer-Programming-based (MIP) optimizers for ML inference queries that makes use of different objectives to find Pareto-optimal inference query plans.
AB - Given a set of pre-trained Machine Learning (ML) models, can we solve complex analytic tasks that make use of those models by formulating ML inference queries? Can we mitigate different tradeoffs, e.g., high accuracy, low execution costs and memory footprint, when optimizing the queries? In this work we present different multi-objective ML inference query optimization strategies, and compare them on their usability, applicability, and complexity. We formulate Mixed-Integer-Programming-based (MIP) optimizers for ML inference queries that makes use of different objectives to find Pareto-optimal inference query plans.
UR - http://www.scopus.com/inward/record.url?scp=85163841187&partnerID=8YFLogxK
U2 - 10.1109/ICDEW58674.2023.00017
DO - 10.1109/ICDEW58674.2023.00017
M3 - Conference contribution
AN - SCOPUS:85163841187
T3 - Proceedings - 2023 IEEE 39th International Conference on Data Engineering Workshops, ICDEW 2023
SP - 74
EP - 78
BT - Proceedings - 2023 IEEE 39th International Conference on Data Engineering Workshops, ICDEW 2023
PB - Institute of Electrical and Electronics Engineers (IEEE)
T2 - 39th IEEE International Conference on Data Engineering Workshops, ICDEW 2023
Y2 - 3 April 2023 through 7 April 2023
ER -