In this preliminary round of the AI Modelling Lab’s 3-minute thesis/project competition, a second batch of speakers will give us an exciting overview of their research with the help of a single slide. The topics range from healthcare/medicine to the formal foundations of physics, covering both knowledge/symbolic- and data-driven AI.
Speakers: Filip Smola, Rui Zhao, James Vaughan, Lauren DeLong, Jake Palmer, Mark Chevallier, Richard Schmoetten, Petros Papapanagiotou
In this preliminary round of the AI Modelling Lab's 3-minute thesis/project competition, a first batch of speakers will give us an exciting overview of their research with the help of a single slide. The topics range from healthcare to formalised mathematics, covering both knowledge/symbolic- and data-driven AI.
Speakers: Imogen Morris, Adarsh Prabhakaran,Jorge Gaete-Villegas, Jiawei Zheng, Zonglin Ji, Ramon Mir Fernández
We are planning to hold a 3 minute thesis competition within the AIML lab In brief: you have 3 minutes and 1 slide to present your thesis at an accessible level. On Friday I will give a brief description of the competition.
Speaker: Imogen Morris
In this talk I will present the use of the Stochastic Block Model (SBM) as an alternative to clustering techniques to study the impact of multimorbidity in in-hospital sepsis and mortality.
Speaker: Jorge Gaete Villegas
In this work, a Relational Graph Attention Network is extended to operate on multimodal biological input and is compared alongside previously established side effect prediction methods to evaluate the efficacy of deep Network Representation Learning for adverse drug reaction prediction.
Speaker: Lauren DeLong
Convex optimisation is a subfield of mathematics that studies convex functions and their maxima/minima over a given domain, with applications in control synthesis, signal processing and operations research to mention a few. We describe the how the Lean theorem prover might be used to rigorously check algorithms in the domain, with neural network verification as a potential case study.
Speaker: Ramon Mir Fernandez
I will talk about a formalised model of Markov Decision Processes with rewards. I'll discuss formal proofs of value convergence and of the existence of an optimal policy by vector representation of functions on the states.
Speaker: Mark Chevallier
I will discuss what we have done in the past year to apply network theory tools to improve the capabilities of interactive theorem provers.
Speaker: James Vaughan
Tree search algorithms are used to solve challenging problems like planning problems and combinatorial optimization search. In this presentation, we shall explore how we can integrate human experience directly in tree search without encoding them into objective functions. This approach aims to address the issues of information loss that lead to degraded solution quality and result in increased computational costs.
Spealer: Mustafa Faisal Abdelwahed
In this project, we look to apply causal inference techniques to purely observational data to investigate the effectiveness of subdural drainage in preventing the recurrence of chronic subdural hematoma.
Spealer: Callum Abott