AI programs, built using large language models, make it possible to
auto...
When machine learning models are trained continually on a sequence of ta...
The goal of this thesis is to develop the optimisation and generalisatio...
Do neural networks generalise because of bias in the functions returned ...
We study the problem of sparse nonlinear model recovery of high dimensio...
How much information does a learning algorithm extract from the training...
Compositionality is a basic structural feature of both biological and
ar...
How far apart are two neural networks? This is a foundational question i...
Training neural networks on large datasets can be accelerated by distrib...
Neural networks are known to be vulnerable to adversarial examples. Care...
Training large neural networks requires distributing learning across mul...