Andrew Holliday

I am a machine learning and robotics researcher and recent PhD graduate. I defended my doctoral thesis in December 2024 at McGill University, where I was supervised by professors Gregory Dudek and Ahmed El-Geneidy. My thesis concerned the use of graph neural nets and reinforcement learning to plan urban transit networks. Prior to the PhD, I completed an M.Sc in computer science under Gregory Dudek, in which I developed a novel type of visual feature for scene recognition and localization over long distances. I have also spent time doing computer vision research at Samsung’s Montreal AI lab, and at the National Institute of Informatics in Tokyo under professor Helmut Prendinger, where I led a team of researchers in researching and publishing a journal paper on semantic segmentation.

My research interests include computer vision, urban planning, reinforcement learning, field robotics, graph neural networks, AI alignment, and neural learning in general. Beyond my work in academia, I have several years of industrial experience in software development, machine learning engineering, and data science, as well as experience leading small research teams in academia.

When not engaged in research, I enjoy spending time outdoors, exploring the city of Montreal, and playing board games with friends. I’m also an avid reader of works of fiction, history, and philosophy, especially ethics and the philosophy of mind.

Selected publications

TM-B

Learning heuristics for transit network design and improvement with deep reinforcement learning

Andrew Holliday, Ahmed El-Geneidy, and Gregory Dudek

Transportmetrica B: Transport Dynamics, 2025

Bib PDF Code

@article{Holliday31122025,
  author = {Holliday, Andrew and El-Geneidy, Ahmed and Dudek, Gregory},
  title = {Learning heuristics for transit network design and improvement with deep reinforcement learning},
  journal = {Transportmetrica B: Transport Dynamics},
  volume = {13},
  number = {1},
  pages = {2561863},
  year = {2025},
  publisher = {Taylor \& Francis},
  doi = {10.1080/21680566.2025.2561863},
  url = {https://doi.org/10.1080/21680566.2025.2561863},
  eprint = {https://doi.org/10.1080/21680566.2025.2561863},
}

AR
Scale-invariant localization using quasi-semantic object landmarks

Andrew Holliday, and Gregory Dudek

Autonomous Robots, 2021

Abs Bib PDF Code

This work presents Object Landmarks, a new type of visual feature designed for visual localization over major changes in distance and scale. An Object Landmark consists of a bounding box b defining an object, a descriptor q of that object produced by a Convolutional Neural Network, and a set of classical point features within b. We evaluate Object Landmarks on visual odometry and place-recognition tasks, and compare them against several modern approaches. We find that Object Landmarks enable superior localization over major scale changes, reducing error by as much as 18% and increasing robustness to failure by as much as 80% versus the state-of-the-art. They allow localization under scale change factors up to 6, where state-of-the-art approaches break down at factors of 3 or more.
@article{holliday2021scale, title = {Scale-invariant localization using quasi-semantic object landmarks}, author = {Holliday, Andrew and Dudek, Gregory}, journal = {Autonomous Robots}, volume = {45}, number = {3}, pages = {407--420}, year = {2021}, publisher = {Springer}, }
CVIU
Speedup of deep learning ensembles for semantic segmentation using a model compression technique

Andrew Holliday, Mohammadamin Barekatain, Johannes Laurmaa, and 2 more authors

Computer Vision and Image Understanding, 2017

Deep Learning for Computer Vision

Abs Bib PDF

Deep Learning (DL) has been proven as a powerful recognition method as evidenced by its success in recent computer vision competitions. The most accurate results have been obtained by ensembles of DL models that pool their results. However, such ensembles are computationally costly, making them inapplicable to real-time applications. In this paper, we apply model compression techniques to the problem of semantic segmentation, which is one of the most challenging problems in computer vision. Our results suggest that compressed models can approach the accuracy of full ensembles on this task, combining the diverse strengths of networks of very different architectures, while maintaining real-time performance.
@article{HOLLIDAY201716, title = {Speedup of deep learning ensembles for semantic segmentation using a model compression technique}, journal = {Computer Vision and Image Understanding}, volume = {164}, pages = {16-26}, year = {2017}, note = {Deep Learning for Computer Vision}, issn = {1077-3142}, doi = {https://doi.org/10.1016/j.cviu.2017.05.004}, url = {https://www.sciencedirect.com/science/article/pii/S1077314217300826}, author = {Holliday, Andrew and Barekatain, Mohammadamin and Laurmaa, Johannes and Kandaswamy, Chetak and Prendinger, Helmut}, keywords = {Semantic segmentation, Model compression, Transfer learning, Real-time application}, }