google-deepmind / dks
Multi-framework implementation of Deep Kernel Shaping and Tailored Activation Transformations, which are methods that modify neural network models (and their initializations) to make them easier to train.
☆57Updated last month
Related projects: ⓘ
- A selection of neural network models ported from torchvision for JAX & Flax.☆44Updated 3 years ago
- Pytorch implementation of preconditioned stochastic gradient descent (affine group preconditioner, low-rank approximation preconditioner …☆105Updated this week
- Meta-learning inductive biases in the form of useful conserved quantities.☆37Updated last year
- Implementations and checkpoints for ResNet, Wide ResNet, ResNeXt, ResNet-D, and ResNeSt in JAX (Flax).☆103Updated 2 years ago
- ☆104Updated last week
- ☆56Updated 2 years ago
- Open source code for EigenGame.☆28Updated last year
- A GPT, made only of MLPs, in Jax☆55Updated 3 years ago
- A collection of optimizers, some arcane others well known, for Flax.☆29Updated 3 years ago
- JAX implementation of Learning to learn by gradient descent by gradient descent☆25Updated 2 years ago
- Image augmentation library for Jax☆36Updated 5 months ago
- 👑 Pytorch code for the Nero optimiser.☆20Updated last year
- Implementation of deep implicit attention in PyTorch☆63Updated 3 years ago
- ☆64Updated 10 months ago
- Experiment of using Tangent to autodiff triton☆66Updated 7 months ago
- A minimal implementation of a VAE with BinConcrete (relaxed Bernoulli) latent distribution in TensorFlow.☆21Updated 4 years ago
- Differentiable Algorithms and Algorithmic Supervision.☆101Updated last year
- Official repository for the paper "Can You Learn an Algorithm? Generalizing from Easy to Hard Problems with Recurrent Networks"☆58Updated 2 years ago
- Very deep VAEs in JAX/Flax☆45Updated 3 years ago
- minGPT in JAX☆45Updated 2 years ago
- Riemannian Convex Potential Maps☆68Updated last year
- Official repository for our ICLR 2021 paper Evaluating the Disentanglement of Deep Generative Models with Manifold Topology☆36Updated 3 years ago
- Tensor Parallelism with JAX + Shard Map☆10Updated 11 months ago
- ☆35Updated 2 years ago
- Texture mapping with variational auto-encoders☆40Updated 2 years ago
- ☆191Updated 4 months ago
- Lightning-like training API for JAX with Flax☆28Updated 4 months ago
- Code accompanying our paper "Feature Learning in Infinite-Width Neural Networks" (https://arxiv.org/abs/2011.14522)☆57Updated 3 years ago
- Usable implementation of Emerging Symbol Binding Network (ESBN), in Pytorch☆23Updated 3 years ago
- Optimization algorithm which fits a ResNet to CIFAR-10 5x faster than SGD / Adam (with terrible generalization)☆11Updated 11 months ago