Projection into the typical set: PITS
A new approach to solving inverse problems
The behaviour of neural flows
Neural nets can struggle to learn very simple flows.
Automatic integration
Like automatic differentiation, but for integration
Requests for research
Some ideas from my masters.
The advantages of backward reasoning.
A simple exploration of what can be gained by reasoning backwards from your goal.
Unsupervised skip connections
How can we use ladder nets in the unsupervised setting?
Visualising dataset alignment
Dataset alignment represented as tensor networks
Conserved complexity
A conservation law for algorithms?
A functional type of non-linear algebra?
Functional programming plus neural network architectures.