MSc Computer Science (Pass with distinction), University of
Pretoria
Class of 2023
My master's degree focused on Artificial Intelligence and Machine Learning (AL/ML). I authored a
dissertation where I made use of Genetic Algorithms (GAs) and
Generic Programming (GP) to automate the design process of a convolutional neural network (CNN).
Additionally, I introduced a variant of Genetic Programming known as Structure-based Genetic Programming
(SBGP), which proved to be more effective for the same task, surpassing both GAs and traditional GP.
Fortunately, this research led to the publication of one conference paper and one journal article. I owe
much of my success to my supervisor, Prof. Nelishia Pillay, whose invaluable guidance contributed
significantly to achieving these results. Prof. Pillay also co-authored the paper and article:
-
Rahul Kapoor and Nelishia Pillay. 2023. Iterative Structure-Based Genetic Programming for Neural
Architecture Search. In Proceedings of the Companion Conference on Genetic and Evolutionary
Computation (GECCO '23 Companion). Association for Computing Machinery, New York, NY,
USA, 595-598. https://doi.org/10.1145/3583133.3590759.
-
Kapoor, R., Pillay, N. A genetic programming approach to the automated design of CNN models for
image classification and video shorts creation. Genet Program Evolvable Mach 25, 10
(2024). https://doi.org/10.1007/s10710-024-09483-5.
BScHons Computer Science, University of Pretoria
Class of 2021
My honour's degree served to consolidate and enhance my foundational knowledge in computer science and
software engineering acquired during my undergraduate studies. The subjects I pursued included
Artificial Intelligence, Image Processing, Data Mining, Digital Forensics, and Computer and Information
Security. As part of the curriculum, I undertook a research project from a range of topics. I researched
machine translation using transformer models, focusing on developing a miniature translation program
capable of translating English into Afrikaans and isiZulu. This involved making use of publicly
accessible parallel corpora (datasets) to train the translation program.
BSc Computer Science, University of Pretoria
Class of 2020
During my undergraduate degree, I had the privilege of learning a multitude of skills and techniques from
exceptionally smart teachers. In my final year, I took subjects such as Software Engineering,
Artificial Intelligence, Computer Networks, and Database Systems. Additionally, I focused on analytical
subjects like Stochastic Processes, Multivariate Analysis, and Time-Series Analysis.
As part of my degree requirements, I undertook a final year project that required me, as part of a team,
to build a piece of software for a corporate client. I was the team leader, and together, my team and I
built a web application using Reinforcement Learning (an AI technique that allows a program to adjust
its parameters over time given new information, i.e., learn over time) to control traffic lights in a
simulated environment. This environment was based on a real stretch of road with multiple intersections.
Data, such as car numbers, car types, and frequency of cars during regular vs. peak hours, was collected
for initial training and randomly generated thereafter. This application, called LightBot, was
well-received by our lecturers, earning me a distinction for the course.
In my earlier
years, I concentrated on core computer science principles such as data structures and algorithms,
concurrent programming, and computer organization and architecture, which provided me with a solid
foundation. In my first year, I also took physics and business management as minors.
Class of 2015
I graduated from high school with seven distinctions and received academic honors in my final year. I
took the following subjects in high school: English Home Language, Afrikaans First Additional Language,
Mathematics, AP Mathematics, Information Technology (Computer Programming), Life Sciences (Biology),
Physical Sciences (Physics and Chemistry), and Accounting.