Bio
Boyko is a Data Scientist within the ARC, working on the research and application of machine learning algorithms. He gained his MSc from the University of Heidelberg. His thesis was on the topic of developing a novel segmentation and tracking algorithm for cells in video microscopy, using representation learning with deep recurrent neural networks.
Before joining the Turing, Boyko worked as a research assistant, supervised by Prof Neill Campbell at the University of Bath and Prof Pietro Cicuta at the University of Cambridge. His research was on automating microscopy tasks, such as malaria diagnostics, using out of distribution methods and variational autoencoders in low data scenarios.
Research interests
Boyko's current work is on researching unsupervised learning methods in the field of machine learning on graphs.
Achievements and awards
The deep learning algorithm developed during his MSc thesis was among the top scoring methods on the MICCAI Cell Tracking Challenge.
His work on using variational autoencoders for out of distribution detection of malaria parasites was presented at MIDL 2022.