arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2305.00449cs.LG
收藏

机器学习算法的可预测性及相关特征提取技术

Predictability of Machine Learning Algorithms and Related Feature Extraction Techniques

Department of Mathematics and Computer Science(数学与计算机科学系)

Yunbo Dong

AI总结:

该研究设计基于矩阵分解的预测系统,针对五十余个openml数据集,探究随机森林、XGBoost、MLP三类算法的性能可预测性,得出三类相关预测结论。

摘要:

本文设计了一个基于矩阵分解的预测系统,用于预测特定模型在特定数据集上的分类准确率。我们对从openml网站收集的五十多个数据集开展了全面的实证研究,探究随机森林、XGBoost和多层感知器(MultiLayer Perceptron, MLP)三种基础机器学习算法的性能预测。具体得到以下结果:1. 利用粗调变体预测微调模型的可预测性;2. 利用特征提取技术预测MLP的可预测性;3. 利用隐式反馈预测模型性能。

英文摘要:

This thesis designs a prediction system based on matrix factorization to predict the classification accuracy of a specific model on a particular dataset. In this thesis, we conduct comprehensive empirical research on more than fifty datasets that we collected from the openml website. We study the performance prediction of three fundamental machine learning algorithms, namely, random forest, XGBoost, and MultiLayer Perceptron(MLP). In particular, we obtain the following results: 1. Predictability of fine-tuned models using coarse-tuned variants. 2. Predictability of MLP using feature extraction techniques. 3. Predict model performance using implicit feedback.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑