Dr Aaron Bostrom
Aaron is Lecturer: Computer Science on our BSc (Hons) Games Development course.
Computer Science is at the core of Games Development. Software development governs the tools, gameplay and development of games.
With an industry background in Games Programming, and a PhD in Machine Learning Dr. Bostrom delivers subject specific knowledge across a wide area of Games Development. This ranges from low level bit manipulation all the way to high level Artificial Intelligence.
My research career is focused on machine learning across a broad range of topics. During my PhD I devised faster and more efficient algorithms for time series classification. This work was extended from univariate signals into the multivariate domain.
Since then I have been researching machine learning and image classification algorithms for aerial image data in bioscience and crop analysis, having spent time in 2018 working in Nanjing, China developing aerial image algorithms for managing mega farms. More recently, I am considering the cross over between Games and machine learning. Specifically how to use machine learning algorithms in real time, whether this is to guide the actions of enemies, or to detect complex gestures/interaction from new human computer interaction technologies.
2018 – What is cost-efficient phenotyping? Optimizing costs for different scenarios – Plant Science
2018 – Shapelet Transforms for Univariate and Multivariate Time Series Classification – University East Anglia (PhD)
2017 – The great time series classification bake off: a review and experimental evaluation of recent algorithmic advances – Data Mining and Knowledge Discovery 31 (3)
2017 – Binary shapelet transform for multiclass time series classification – Transactions on Large-Scale Data-and Knowledge-Centered Systems XXXII
2016 – Evaluating Improvements to the Shapelet Transform – MiLeTS workshop (Knowledge and Data Discovery)
2015 – Binary shapelet transform for multiclass time series classification – International Conference on Big Data Analytics and Knowledge Discovery
2015 – Time-series classification with COTE: the collective of transformation-based ensembles – IEEE Transactions on Knowledge and Data Engineering 27 (9)
2014 – Facility 47
2013 – Mahjong : The Secret Garden
2013 – Jigsaw World
2013 – Just Escape
2013 – Escape from Darkmoor Manor