Arinbjörn Kolbeinsson
My focus is on advancing research and transforming complex ideas into practical solutions, using machine learning, engineering and statistics to drive progress in healthcare, biomedicine and law. I’m passionate about building from scratch, tackling challenges and delivering meaningful outcomes.
-
PhD: Deep learning for health outcome prediction
Imperial College London, 2021 -
MRes in Biomedical Research
Imperial College London, 2016 -
MEng in Bioengineering
Imperial College London, 2015 -
BA in Law
University of Iceland, 2024
- Personalised health outcome prediction and action optimisation
- Time-series analysis to identify changes or anomalies in temporal data
- Multi-modal learning
- Tensor methods for improving robustness and generalisability
I am currently a visiting scholar at the University of Virginia where I collaborate with Tom Hartvigsen on advancing the efficiency, computational speed, and ecological footprint of Large Language Models (LLMs). In this capacity, I am also contributing to the development of cutting-edge model editing techniques.
At Evidation Health I developed machine learning methods to predict health outcomes based on high-frequency multi-modal data. I also
collaborated on research with academic institutions on differential privacy in machine learning.
Read more on evidation.com
During my PhD at Imperial College London, I developed a deep neural network model to
predict brain age difference
using over 20,000 MRI scans from UK Biobank. Through a phenome-wide association analysis, we identified significant
associations between this learnt biomarker and over 40 traits, including cognitive diseases and poorer cognitive function, highlighting
relationships and possible causation between brain and systemic health.
Central to my research was developing new machine learning models to analyse these complex health data.
I created
a method to learn features from 3D brain volumes
(open access version)
with a stochastic tensor regression layer to better
leverage the structural information in the images. Part of this work was done during my internship at Samsung AI, in
collaboration with Professor Anima Anandkumar’s group at Caltech.
- Arinbjörn Kolbeinsson, Kyle O'Brien, Tianjin Huang, Shanghua Gao, Shiwei Liu, Jonathan Richard Schwarz, Anurag Vaidya, Faisal Mahmood, Marinka Zitnik, Tianlong Chen and Thomas Hartvigsen. "Composable Interventions for Language Models". Pre-print (2024).
- Mike A Merrill, Esteban Safranchik, Arinbjörn Kolbeinsson, Piyusha Gade, Ernesto Ramirez, Ludwig Schmidt, Luca Foschini, Tim Althoff. "Homekit2020: A benchmark for time series classification on a large mobile sensing dataset with laboratory tested ground truth of influenza infections". Conference on Health, Inference, and Learning (CHIL) (2023).
- Georgios Efstathiadis, Patrick Emedom-Nnamdi, Arinbjörn Kolbeinsson, Jukka-Pekka Onnela, Junwei Lu. "STASIS: Reinforcement Learning Simulators for Human-Centric Real-World Environments" Trustworthy Machine Learning for Healthcare (2023).
- Arinbjörn Kolbeinsson and Luca Foschini. "Generative models for wearables data." Workshop on Deep Generative Models for Health at NeurIPS (2023).
- Arinbjörn Kolbeinsson, Naman Shukla, Akhil Gupta, Lavanya Marla, and Kartik Yellepeddi. "Galactic Air Improves Ancillary Revenues with Dynamic Personalized Pricing." INFORMS Journal on Applied Analytics (2022).
- Arinbjörn Kolbeinsson, Jean Kossaifi, Yannis Panagakis, Adrian Bulat, Anima Anandkumar, Ioanna Tzoulaki, and Paul Matthews. "Tensor Dropout for Robust Learning," in IEEE Journal of Selected Topics in Signal Processing, doi: 10.1109/JSTSP.2021.3064182 (2021).
- Akhil Gupta, Lavanya Marla, Ruoyu Sun, Naman Shukla and Arinbjörn Kolbeinsson. “PenDer: Incorporating Shape Constraints via Penalized Derivatives”. In Proceedings of the AAAI Conference on Artificial Intelligence (2021).
- Arinbjörn Kolbeinsson, Sarah Filippi, Yannis Panagakis, Paul M. Matthews, Paul Elliott, Abbas Dehghan, and Ioanna Tzoulaki. "Accelerated MRI-predicted brain ageing and its associations with cardiometabolic and brain disorders." Scientific Reports 10, no. 1 (2020).
- Kossaifi, Jean, Zachary C. Lipton, Arinbjörn Kolbeinsson, Aran Khanna, Tommaso Furlanello, and Anima Anandkumar. "Tensor regression networks." Journal of Machine Learning Research 21, no. 123 (2020).
- See all on my Google Scholar profile.