no code implementations • 17 Dec 2019 • Daniel C. Castro, Ian Walker, Ben Glocker
This article discusses how the language of causality can shed new light on the major challenges in machine learning for medical imaging: 1) data scarcity, which is the limited availability of high-quality annotations, and 2) data mismatch, whereby a trained algorithm may fail to generalize in clinical practice.
no code implementations • 14 May 2019 • Ian Walker, Ben Glocker
We propose a novel Bayesian nonparametric method to learn translation-invariant relationships on non-Euclidean domains.
no code implementations • ICML 2018 • Konstantinos Kamnitsas, Daniel C. Castro, Loic Le Folgoc, Ian Walker, Ryutaro Tanno, Daniel Rueckert, Ben Glocker, Antonio Criminisi, Aditya Nori
We present a novel cost function for semi-supervised learning of neural networks that encourages compact clustering of the latent space to facilitate separation.
no code implementations • 2 Aug 2017 • Bryan Willimon, Steven Hickson, Ian Walker, Stan Birchfield
In particular, we show that our method is able to estimate the configuration of a textureless nonrigid object with no correspondences available.