PhD Thesis

• Bayesian Optimisation & Priors:
  Improving Air Pollution Monitoring and Hyperparameter Optimisation
  Sigrid Passano Hellan, University of Edinburgh, 2023
  Supervised by Nigel Goddard and Christopher Lucas.

Publications

• Hyperparameter Selection in Continual Learning
  Thomas L. Lee, Sigrid Passano Hellan, Linus Ericsson, Elliot J. Crowley and Amos Storkey
  Under review.
• Data-driven Prior Learning for Bayesian Optimisation
  Sigrid Passano Hellan, Christopher Lucas and Nigel Goddard
  Presented at the Adaptive Experimental Design and Active Learning in the Real World workshop at NeurIPS 2023. Code is here.
• Obeying the Order: Introducing Ordered Transfer Hyperparameter Optimisation
  Sigrid Passano Hellan, Huibin Shen, François-Xavier Aubet, David Salinas and Aaron Klein
  Presented at the AutoML 2023 conference workshop. Code is here.
• Bayesian Optimisation Against Climate Change: Applications and Benchmarks
  Sigrid Passano Hellan, Christopher Lucas and Nigel Goddard
  Presented at the Data-centric Machine Learning Research workshop at ICML 2023. Code is here.
• Bayesian Optimisation for Active Monitoring of Air Pollution
  Sigrid Passano Hellan, Christopher Lucas and Nigel Goddard
  Presented at AAAI 2022. Code available here. Poster available here.
• Optimising Placement of Pollution Sensors in Windy Environments
  Sigrid Passano Hellan, Christopher Lucas and Nigel Goddard
  Presented at the AI for Earth Sciences Workshop at NeurIPS 2020. Check out the video here.

Tutorials

• Using Syne Tune for Transfer Learning contributed to the official Syne Tune documentation.
• Derivatives and Gaussian Processes published on the CDT in Data Science blog.

Talks

• Affordable Bayesian Optimisation Contributed talk at the First Edinburgh Workshop on Affordable Machine Learning.

Blog posts

• Spotting an Air Pollution Sensor in the Wild, 12 May 23.

Google Scholar profile.