Hi! My name is Amin Ravanbakhsh, and I am a Master's student at the University of Waterloo. As a member of the Data Analytics Lab, I am fortunate to be supervised by Professor Ali Ghodsi.
I am broadly interested in the problem of symbolic regression and how large language models could be used to solve this problem. My research focuses on developing a reasoning-based tool that leverages human guidance to solve symbolic regression.
I am also interested in Cybersecurity, Generative AI, Reinforcement learning and Stochastic Processes.
Master of Science in Computer Science
May 2023 - present
Bachelor of Science in Computer Engineering
Sep 2017 - Jul 2022
Member of Iran’s Physics Olympiad team
Jun 2016 - Jul 2017
International Physics Olympiad (IPHO) Gold Medalist
Jul 2017
Iran’s National Physics Olympiad Gold Medalist
Jul 2016
• Developed an AI-based End-to-End Cybersecurity Platform prototype to identify and analyze adversarial
techniques, providing a broad overview of events to enhance security defenses.
• Integrated the Malware Information Sharing Platform (MISP) project as a knowledge set along with
Retrieval-Augmented Generation (RAG), Large Language Models, and Search Engine Indexing to
identify cyber attacks.
Jun. 2024 - Aug. 2024 , Waterloo, Canada
• Enhanced Cylance AI, Cybersecurity End-to-End platform, by advancement of performance in Clustering
group of adversarial attacks.
• Leveraged Data Bricks as a Big Data technology to boost threat detection and hunting, providing insights
that strengthened security and informed effective response strategies.
Jan. 2024 - Jun. 2024 , Waterloo, Canada
• Collaborated with the Threat Hunting team as a Machine Learning Engineer.
• Developing a reasoning-based Symbolic Regression tool that leverages Large Language Models along with metadata and axioms as a knowledge set to identify mathematical equations describing tabular datasets.
• Utilizing and advancing Symbolic GPT to identify interpretable equations from Physics datasets under the mentorship of Professor Ali Ghodsi, driving significant progress in the field of symbolic regression.
Jan. 2023 - Now , Waterloo, Canada
• Employed Bayesian Inference in conjunction with Thompson sampling to address the Multi-armed Bandit problem through Reinforcement Learning.
• Conducted a comprehensive survey of Bayesian algorithms to determine the most suitable algorithm for designing a recommendation system based on industry-specific data.
Sep. 2021 - Jun. 2022 , Tehran, Iran
• Developed a robust system resistant to unexpected data changes (Concept Drift) by utilizing Attentive Aggregation within Federated Learning, with applications in the Internet of Vehicles.
• Conducted empirical tests on the attentive model as part of the research team.
Jun. 2021 - Apr. 2022 , Montreal, Remote
Comparison of factorization machines method with common methods for classification and clustering about categorical data. Implementation of a recommendation system for YektaNet company’s merchandise. Medium Link
Employed the VGG16 network for detecting tumors in brain images, and utilized the Grad-CAM algorithm to visualize the underlying reasons for VGG16's malignant tumor detection. Code
Classified ECG time series data using LSTM and CNN networks, and conducted a comparative analysis to highlight the advantages of LSTM over CNN. Code
Implemented a Generative Adversarial Networks (GANs) to generate artificial digit images that closely resemble real handwritten digits. Code
Designed and implemented a driver drowsiness detection system, leveraging a neural network to analyze facial expressions and issue warnings. Successfully deployed the project on an Arduino board. Code
Implemented a recommendation system using movie synopsis. The system includes a search engine that employs the TF-IDF (Term Frequency - Inverse Document Frequency) algorithm to find movies related to specific search terms. Furthermore, a Gaussian Mixture model is utilized to categorize movies into distinct clusters. Code
CE 401717: Machine Learning (Graduate course) - Spring 2022, Fall 2021
CE 40951.5: Intelligent Analysis of Biomedical Images (Graduate course) - Spring 2022
CE 40417: Artificial Intelligence - Fall 2021
CE 40181: Probability and Statistics - Fall 2020