Continual Learning in Machine Intelligence: A Comparative Analysis of Model Performance

Author:

Gajjar Kimi1,Choksi Ami2,Gajjar T.3

Affiliation:

1. Elite Technocrats

2. C. K. Pithawala College of Engineering and Technology

3. Western Sydney University

Abstract

Abstract

Continual Learning (CL) is crucial in artificial intelligence for systems to maintain relevance and effectiveness by adapting to new data while retaining previously acquired knowledge. This study explores the performance of multiple machine learning algorithms in CL tasks across various stock symbol datasets over different years. The algorithms assessed include decision trees, ridge regression, lasso regression, elastic net regression, random forests, support vector machines, gradient boosting, and Long Short-Term Memory (LSTM). These models are evaluated on their ability to incrementally gather and maintain knowledge over time, crucial for continual learning. Performance is measured using Mean Squared Error (MSE) and R-squared metrics to assess predictive precision and data conformity. Additionally, the evaluation extends to consider stability, flexibility, and scalability—important factors for models operating in dynamic environments. This comprehensive analysis aims to identify which algorithms best support the objectives of continual learning by effectively integrating new information without compromising the integrity of existing knowledge.

Publisher

Research Square Platform LLC

Reference38 articles.

1. How to reuse and compose knowledge for a lifetime of tasks: A survey on continual learning and functional composition;Mendez JA;arXiv preprint arXiv,2022

2. Wang, Z., Liu, L., Duan, Y., & Tao, D. (2022, June). Continual learning through retrieval and imagination. In Proceedings of the AAAI Conference on Artificial Intelligence (Vol. 36, No. 8, pp. 8594–8602).

3. DataStory™: an interactive sequential art approach for data science and artificial intelligence learning experiences;Maxwell D;Innovation and Education,2021

4. Inference and Learning in a Latent Variable Model for Beta Distributed Interval Data;Mousavi H;Entropy,2021

5. Opposition-based sine cosine optimizer utilizing refraction learning and variable neighborhood search for feature selection;Abed-Alguni BH;Applied Intelligence,2023

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3