Showing 2024Show all
3 papers · 1 filter
cs.LG2024
Neural Network Symmetrisation in Concrete Settings
Rob Cornish
Cornish (2024) recently gave a general theory of neural network symmetrisation in the abstract context of Markov categories. We give a high-level overview of these results, and the…
cs.LG2024
SymDiff: Equivariant Diffusion via Stochastic Symmetrisation
Leo Zhang, Kianoosh Ashouritaklimi, Yee Whye Teh +1
We propose SymDiff, a method for constructing equivariant diffusion models using the framework of stochastic symmetrisation. SymDiff resembles a learned data augmentation that is d…
stat.ML2024
Stochastic Neural Network Symmetrisation in Markov Categories
Rob Cornish
We consider the problem of symmetrising a neural network along a group homomorphism: given a homomorphism , we would like a procedure that converts -equivariant neur…