Mathematical models are used to inform decisions across many sectors including climate change, finance, and epidemics. But as Erica Thompson explains in this Oxford Mathematics Public Lecture, models are not perfect representations of the real world – they are partial, uncertain and often biased. What, then, does responsible modelling look like? And how can we apply this ethical framework to new AI modelling methods?
Alexandre Tsybakov (CREST, ENSAE, Institut Polytechnique de Paris and pictured) - Gradient-free stochastic optimization
3pm, June 6th, L1, Andrew Wiles Building
Full details plus a 'Distinguished Speaker' seminar on Bayesian Probabilistic Subnational Population Projections
Editorial: Capillarity and elastocapillarity in biology
Hu, D
Kim, H
Vella, D
Interface Focus
volume 15
issue 2
20250020
(16 May 2025)
Oxford Mathematics Public Lecture - Wednesday 4 June, 5pm, L1
Mathematical models are used to inform decisions across many sectors including climate change, finance, and epidemics. But models are not perfect representations of the real world – they are partial, uncertain and often biased. What, then, does responsible modelling look like? And how can we apply this ethical framework to new AI modelling methods?
The College Store, who sell our stash, including the new Oxford Mathematics Backpack, are running a discount week (a ten day week) from today Friday 23rd May to Sunday 1st June.
Just click here and use the code TRINITY10.
Dynamic Sparse No Training: Training-Free Fine-tuning for Sparse LLMs
Zhang, Y
Zhao, L
Lin, M
Sun, Y
Yao, Y
Han, X
Tanner, J
Liu, S
Ji, R
(13 Oct 2023)
http://arxiv.org/abs/2310.08915v3
Oxford Mathematician Mike Giles is a computational mathematician who has worked at the interface with both engineering and computer science. His early research was on computational fluid dynamics, developing algorithms and software which is today used by Rolls-Royce in the design of its aircraft engines. More recently, he moved into computational finance and more generally the area of Uncertainty Quantification, developing advanced Monte Carlo simulation methods