Develop an Innovative Particle Swarm Algorithm for Solving the Nonlinear Constrained Optimization Problems and Automatic Search the Hyperparameters of Machine Learning

Lu, Hao-Chun (PI)
Chu, Pao-Hsien (CoPI)

Project: National Science and Technology Council › National Science and Technology Council Academic Grants

Abstract

跟許多國內外學者交流後, 他們在解業界問題常會面臨下列困擾: (1)使用PSO, 到求解後期時無法有效精進問題解。求解非線性約束規劃(2)使用機器學習時, 因沒有搜尋到合適的超參數, 而使成效不彰。成千上萬的實務問題都是NCOPs, 且因人工智慧的崛起, 導致有太多場合需要有效尋找機器學習之最佳超參數。本計畫會產出優質改良PSO以解決上述問題, 不僅對學術是一大貢獻, 對產業界也會造成重大的影響, 同時加強人工智慧的推廣與普及化。

Project IDs

Project ID:PB11007-3541
External Project ID:MOST110-2410-H182-008-MY3

Status	Finished
Effective start/end date	01/08/21 → 31/07/22

Keywords

Particle swarm optimization
Stagnant particle
Machine learning
Hyperparameter
constrained nonlinear programming problems.

Access Project

https://www.grb.gov.tw/search/planDetail?id=13879932

Develop an Innovative Particle Swarm Algorithm for Solving the Nonlinear Constrained Optimization Problems and Automatic Search the Hyperparameters of Machine Learning

Project Details

Abstract

Project IDs

Keywords

Access Project

Fingerprint