Many areas of machine learning and science involve large linear algebra
...
Diffusion models are a class of probabilistic generative models that hav...
Unlike conventional grid and mesh based methods for solving partial
diff...
No free lunch theorems for supervised learning state that no learner can...
While there has been progress in developing non-vacuous generalization b...
Equivariance guarantees that a model's predictions capture key symmetrie...
Physics-inspired neural networks (NNs), such as Hamiltonian or Lagrangia...
There is often a trade-off between building deep learning systems that a...
State-of-the-art methods for scalable Gaussian processes use iterative
a...
Symmetries and equivariance are fundamental to the generalization of neu...
Reasoning about the physical world requires models that are endowed with...
Invariances to translations have imbued convolutional neural networks wi...
Continuous input signals like images and time series that are irregularl...
The translation equivariance of convolutional layers enables convolution...
Normalizing flows transform a latent distribution through an invertible
...
Recent advances in deep unsupervised learning have renewed interest in
s...