Jasper Gerigk

Jasper Gerigk

Undergraduate Student in Computer Science and Mathematics

University of Toronto

I am a fourth year computer science and mathematics student at the University of Toronto. My interest lie in understanding the foundations of deep reinforcement learning, the development of algorithms for situations with sparse and misleading rewards, and utilising novel model structures for non-Euclidean data for better performance and generalisation. In general, how to develop robust agents capable of learning to solve in general complex tasks.

In my work at Bosch, we used graph neural networks and deep reinforcement learning to optimize traffic flow at intersections. We were able to train and evaluate models that outperformed traditional approaches and generalised well to unseen intersection layouts. Building on this experience, I applied geometric deep learning, under the supervision of Professor Engels, to the problem of learning strategies for the deck-building game Dominion. The resulting agent were able to learn strategies previously only seen in human play on a wider set of game configurations.

Currently, I am working on object-oriented reinforcement learning using pre-trained computer vision foundation models in the Toronto Intelligent Systems Lab led by Professor Gilitschenski.

Academic Interests
  • Artificial Intelligence
  • Reinforcement Learning
  • Geometric Deep Learning
Education
  • BSc Computer Science Specialist and Mathematics Major, 2024

    University of Toronto, Canada

  • IB Diploma, 2019

    Metropolitan School Frankfurt, Germany

PUBLICATIONS

EXPERIENCE

Member of the TISL research group led by Professor Gilitschenski
Worked on combining pre-trained computer vision models with reinforcement learning
Methods applied: Pytorch, JAX, Rainbow, SAM, FastSAM
Mercedes Benz AG, Böblingen, Germany
Data Analytics Internship
Member of the Fleet Learning for Automated Driving team
For the first time, used large-scale customer fleet data to analyze lateral vehicle movement to improve comfort of lane following assistant
Methods applied: Spark, Azure Databricks, Frequentist and Bayesian statistics in Python
Member of BMWK-funded research project “Lokales Umfeldmodell für das Kooperative, Automatisierte Fahren in komplexen Verkehrssituationen”
Development of multi-agent reinforcement learning algorithms for centralized planning of connected self-driving vehicles using graph neural networks
Methods applied: DQN, TD3, RCGN, GAT implemented in Python using PyTorch
Excubo AG, Zug, Switzerland
Software Development Internship
Designed and built functional software demonstration based on Server-Side Blazor (C#)
Contributed to backend by integrating machine learning methods using Python
German Research Center for Artificial Intelligence (DFKI), Kaiserslautern, Germany
Student Internship
Created instructional material for AI undergraduate course at TU Kaiserslautern on Reinforcement Learning including Deep-Q learning for Brick Breaker using PyTorch
Fraunhofer Institute for Intelligent Analysis and Information Systems, St. Augustin, Germany
Student Internship
Implemented linear least squares for multiclass classification of geographic coordinates

EDUCATION

University of Toronto
Bachelor of Science
University of Toronto
September 2019 – Present Toronto, Canada

GPA: 3.96/4.0

Programs:

  • Computer Science Specialist
  • Focus In Artificial Intelligence
  • Mathematics Major

Advanced Courses:

  • CSC311: Introduction to Machine Learning
  • CSC324: Principles of Programming Languages
  • CSC412: Probabilistic Learning and Reasoning
  • CSC413: Neural Networks and Deep Learning
  • CSC420: Introduction to Image Understanding
  • CSC494: Computer Science Project: Developed and published agent for Dominion under supervision of Professor Engels

Awards:

  • 2020, 2023 - Dean’s List Scholar
  • 2019 - 2020 - Millard Scholarship ($1208)
  • 2022 - 2023 - Dr. James A. & Connie P. Dickson Scholarship In Science & Mathematics ($500): Recognizes best University College students in science and mathematics
  • 2022 - 2023 - University of Toronto Scholar ($1500): Recognizes the most outstanding students across all three campuses of the University of Toronto
  • 2024 Computing Research Association’s Outstanding Undergraduate Researcher Award Honorable Mention
Technical University of Darmstadt
Bachelor of Science - Exchange
Technical University of Darmstadt
May 2021 – October 2021 Darmstadt, Germany

Courses covering:

  • Statistical Machine Learning
  • Natural Language Processing using Deep Learning
Johannes Gutenberg University Mainz
Bachelor of Science - Exchange
Johannes Gutenberg University Mainz
November 2020 – October 2021 Mainz, Germany

Courses covering:

  • Linear Algebra
  • Probability and Statistics
  • Numerical Methods

PROJECTS

Cosmos
Cosmos
Core-maintainer of the Cosmos open source project
Cosmos provides the foundation for the development of operating systems in C# and provides a custom compiler, standard library, and drivers
LearnAI
LearnAI
Participated in LearnAI program at UofT
Presented final project at StartAI Conference