Student Stories: Joshua Murphy, Cohort 3, Year 2 - From Physics to Statistics!
The Distributed Algorithms CDT at the University of Liverpool engages its students in impactful, cutting-edge and diverse research, allowing them to collaborate and develop their skills within the university’s data science community. Read about Josh's journey…
My Journey
After my bachelors in physics at the University of Warwick, I found myself drawn more to statistics so pursued a master’s in applied statistics at the University of Strathclyde. I was particularly interested in the research elements of my master’s so made the decision quite late in the academic year to look at PhDs. Like many, I conducted my search primarily through FindaPhD which was also how I came across the Distributed Algorithms CDT. I sent in my application in June and had three interviews for the projects that piqued my interest, in July. I received an offer for my top pick shortly after interview which I accepted practically instantly and sent apologetic cancellation emails to my other scheduled interviews. I began my PhD in October 2021 as part of Cohort 3.
My Project
My project title is ‘Scalable Online Machine Learning’. I will mainly focus on taking a model rather than metric-based approach to tackling concept drift. Concept drift occurs when the statistical environment in which a model is trained ends up being substantially different from that in which it is deployed leading to reduced model performance. To tackle this problem, I am using Sequential Monte Carlo (SMC) samplers which are also heavily used for a multitude of applications within the Big Hypotheses sub-group. The project is co-sponsored by GCHQ.
My Experience at the CDT
A common refrain amongst DA CDT students is the benefit of the cohort environment along with exposure to the wider signal processing research processing group. The wealth of people with varied experiences and backgrounds means an expert or at least someone who can go a long way to pointing you in the right direction is never far away. I feel particularly lucky because I started alongside another student with the exact same project title, Andrew Millard. This really ensured that the initial phases of the PhD were far less lonely and directionless than they could have been because our initial paths were very similar. While our projects have diverged significantly in time, he remains an invaluable source of information and advice.
Distributed Algorithms CDT
The Distributed Algorithms CDT is an Innovative Data Science, AI and Machine Learning Research Centre, aligning PhD students, academics and industrialists to work together to generate novel solutions to tough data science challenges. If you would like to find out more about our programme and would like to talk about becoming an active member of our CDT community, please visit our website or email kelli.cassidy@liverpool.ac.uk.
The wealth of people with varied experiences and backgrounds means an expert or at least someone who can go a long way to pointing you in the right direction is never far away.