Hyper-Parameter Optimization
Abstract
The quality of machine learning models largely depends on their hyper-parameter configurations. However, finding proper hyper-parameter configurations for machine learning algorithms/systems is challenging and requires much resources. This project aims to research on novel hyper-parameter optimization techniques to support efficient machine learning systems.
Project members
Zeyi WEN
Assistant Professor
Publications
1. Efficient Hyperparameter Optimization with Adaptive Fidelity Identification. Jiantong Jiang*, Zeyi Wen, Atif Mansoor, and Ajmal Saeed Mian. IEEE/CVPR 2024
2. Enhancing the Performance of Bandit-based Hyperparameter Optimization. Yile Chen, Zeyi Wen*, Jian Chen*, and Jin Huang.
Project Period
2022-Present
Research Area
Data-driven AI & Machine Learning、High-Performance Systems for Data Analytics
Keywords
Hyper-parameter Optimization, Machine Learning, Resource Efficiency