cambridge-mlg / miracle
This repository contains the code for our recent paper `Minimal Random Code Learning: Getting Bits Back from Compressed Model Parameters'
☆21Updated 6 years ago
Related projects ⓘ
Alternatives and complementary repositories for miracle
- Code release for Hoogeboom, Emiel, Jorn WT Peters, Rianne van den Berg, and Max Welling. "Integer Discrete Flows and Lossless Compression…☆97Updated 4 years ago
- Proximal Mean-field for Neural Network Quantization☆22Updated 4 years ago
- Regularization, Neural Network Training Dynamics☆14Updated 4 years ago
- Feasible target propagation code for the paper "Deep Learning as a Mixed Convex-Combinatorial Optimization Problem" by Friesen & Domingos…☆28Updated 6 years ago
- ☆36Updated 3 years ago
- Code accompanying our paper "Finding trainable sparse networks through Neural Tangent Transfer" to be published at ICML-2020.☆13Updated 4 years ago
- This repository is no longer maintained. Check☆82Updated 4 years ago
- Estimating Gradients for Discrete Random Variables by Sampling without Replacement☆39Updated 4 years ago
- Implementation of BinaryConnect on Pytorch☆36Updated 3 years ago
- Monotone operator equilibrium networks☆51Updated 4 years ago
- TensorFlow implementation of "noisy K-FAC" and "noisy EK-FAC".☆60Updated 5 years ago
- Limitations of the Empirical Fisher Approximation☆45Updated 4 years ago
- Demo: Slightly More Bio-Plausible Backprop☆22Updated 7 years ago
- ☆82Updated 4 years ago
- ☆27Updated 6 years ago
- Lookahead: A Far-sighted Alternative of Magnitude-based Pruning (ICLR 2020)☆33Updated 4 years ago
- Piecewise Linear Functions (PWL) implementation in PyTorch☆48Updated 2 years ago
- Experiments for the paper "Exponential expressivity in deep neural networks through transient chaos"☆70Updated 8 years ago
- Successfully training approximations to full-rank matrices for efficiency in deep learning.☆16Updated 3 years ago
- Code for the article "What if Neural Networks had SVDs?", to be presented as a spotlight paper at NeurIPS 2020.☆69Updated 4 months ago
- Implementation of Methods Proposed in Preventing Gradient Attenuation in Lipschitz Constrained Convolutional Networks (NeurIPS 2019)☆33Updated 4 years ago
- ☆15Updated 5 years ago
- Delta Orthogonal Initialization for PyTorch☆18Updated 6 years ago
- Official code for ICML 2020 paper "Variational Bayesian Quantization"☆23Updated last year
- ☆32Updated 6 years ago
- ☆40Updated 5 years ago
- ☆13Updated 5 years ago
- Percentile computation for pytorch☆20Updated 4 years ago
- Code for the paper "Training Binary Neural Networks with Bayesian Learning Rule☆37Updated 2 years ago
- Compression with Flows via Local Bits-Back Coding☆39Updated 5 years ago