Rashid Barket

Rashid is a PhD student in the field of machine learning. He is conducting his research at the Centre for Computational Science and Mathematical Modelling at Coventry University.

Rashid received his bachelors degree in Mathematics and Computing Science at Simon Fraser University (SFU) in Canada. His Research interests are in Machine Learning, Computer Algebra, and NLP

Previous Research Experience:

• Ester Lab - Machine Learning on single-cell data for cell type discovery
• Complex Systems Modelling Group - Building both epidemiological and machine learning models to use on COVID-19 data from British Columbia hospitals. 

Work Experience

• SFU Big Data Hub - Junior Big Data Analyst, 2021 May - 2022 May
• MasterCard - Data Insights Platform intern, 2020 May - 2020 Aug
• Trulioo - Data Analyst, 2018 Sep - 2019 May

Title: Use of machine learning technology to improve symbolic simplification and integration in the Maple computer algebra system, without risking their mathematical correctness.

Abstract: A Computer Algebra System (CAS) is software used for manipulating and performing calculations with mathematical expressions. One key feature of a CAS is its exactness; A CAS will always return a correct answer and never an approximate answer (or no answer if it can't be calculated). A field such as Machine Learning (ML) does not seem likely to pair well with Computer Algebra due to the approximate nature of ML.

Rather than trying to use ML to calcuate the answer to a problem directly (such as integrating a function), we can instead use ML to help guide algorithms that do these calculations to perform better. This is the case of the integration and simplification functions in Maple, one of the top CASs. Integrate and Simplify both have many sub-algorithms to try and make their calculation work. Rather than naively trying each one randomly, we instead will use ML to help pick a sub-algorithm to try, making the computation faster/optimal