• Stars
    star
    451
  • Rank 96,968 (Top 2 %)
  • Language
  • License
    Apache License 2.0
  • Created over 8 years ago
  • Updated about 4 years ago

Reviews

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Repository Details

♾A curated list of papers and code about very deep neural networks



awesome-very-deep-learning is a curated list for papers and code about implementing and training very deep neural networks.

Neural Ordinary Differential Equations

ODE Networks are a kind of continuous-depth neural network. Instead of specifying a discrete sequence of hidden layers, they parameterize the derivative of the hidden state using a neural network. The output of the network is computed using a black-box differential equation solver. These continuous-depth models have constant memory cost, adapt their evaluation strategy to each input, and can explicitly trade numerical precision for speed.

Papers

  • Neural Ordinary Differential Equations (2018) [original code], introduces several ODENets such as continuous-depth residual networks and continuous-time latent variable models. The paper also constructs continuous normalizing flows, a generative model that can train by maximum likelihood, without partitioning or ordering the data dimensions. For training, the authors show how to scalably backpropagate through any ODE solver, without access to its internal operations. This allows end-to-end training of ODEs within larger models. NIPS 2018 best paper.
  • Augmented Neural ODEs (2019), neural ODEs preserve topology, thus their learned flows can't intersect with each other. Therefore some functions can't be learned. Augmented NODEs improve upon this by adding an additional dimension to learn simpler flows.

Implementations

  1. Authors Autograd Implementation

Value Iteration Networks

Value Iteration Networks are very deep networks that have tied weights and perform approximate value iteration. They are used as an internal (model-based) planning module.

Papers

  • Value Iteration Networks (2016) [original code], introduces VINs (Value Iteration Networks). The author shows that one can perform value iteration using iterative usage of convolutions and channel-wise pooling. It is able to generalize better in environments where a network needs to plan. NIPS 2016 best paper.

Densely Connected Convolutional Networks

Densely Connected Convolutional Networks are very deep neural networks consisting of dense blocks. Within dense blocks, each layer receives the feature maps of all preceding layers. This leverages feature reuse and thus substantially reduces the model size (parameters).

Papers

Implementations

  1. Authors' Caffe Implementation
  2. Authors' more memory-efficient Torch Implementation.
  3. Tensorflow Implementation by Yixuan Li.
  4. Tensorflow Implementation by Laurent Mazare.
  5. Lasagne Implementation by Jan Schlüter.
  6. Keras Implementation by tdeboissiere.
  7. Keras Implementation by Roberto de Moura Estevão Filho.
  8. Chainer Implementation by Toshinori Hanya.
  9. Chainer Implementation by Yasunori Kudo.
  10. PyTorch Implementation (including BC structures) by Andreas Veit
  11. PyTorch Implementation

Deep Residual Learning

Deep Residual Networks are a family of extremely deep architectures (up to 1000 layers) showing compelling accuracy and nice convergence behaviors. Instead of learning a new representation at each layer, deep residual networks use identity mappings to learn residuals.

Papers

Implementations

  1. Torch by Facebook AI Research (FAIR), with training code in Torch and pre-trained ResNet-18/34/50/101 models for ImageNet: blog, code
  2. Torch, CIFAR-10, with ResNet-20 to ResNet-110, training code, and curves: code
  3. Lasagne, CIFAR-10, with ResNet-32 and ResNet-56 and training code: code
  4. Neon, CIFAR-10, with pre-trained ResNet-32 to ResNet-110 models, training code, and curves: code
  5. Neon, Preactivation layer implementation: code
  6. Torch, MNIST, 100 layers: blog, code
  7. A winning entry in Kaggle's right whale recognition challenge: blog, code
  8. Neon, Place2 (mini), 40 layers: blog, code
  9. Tensorflow with tflearn, with CIFAR-10 and MNIST: code
  10. Tensorflow with skflow, with MNIST: code
  11. Stochastic dropout in Keras: code
  12. ResNet in Chainer: code
  13. Stochastic dropout in Chainer: code
  14. Wide Residual Networks in Keras: code
  15. ResNet in TensorFlow 0.9+ with pretrained caffe weights: code
  16. ResNet in PyTorch: code
  17. Ladder Network for Semi-Supervised Learning in Keras : code

In addition, this code by Ryan Dahl helps to convert the pre-trained models to TensorFlow.

Highway Networks

Highway Networks take inspiration from Long Short Term Memory (LSTM) and allow training of deep, efficient networks (with hundreds of layers) with conventional gradient-based methods

Papers

Implementations

  1. Lasagne: code
  2. Caffe: code
  3. Torch: code
  4. Tensorflow: blog, code
  5. PyTorch: code

Very Deep Learning Theory

Theories in very deep learning concentrate on the ideas that very deep networks with skip connections are able to efficiently approximate recurrent computations (similar to the recurrent connections in the visual cortex) or are actually exponential ensembles of shallow networks

Papers

More Repositories

1

awful-ai

😈Awful AI is a curated list to track current scary usages of AI - hoping to raise awareness
6,797
star
2

deep-learning-book

📖 MIT Deep Learning Book in PDF format
542
star
3

spatial-transformer-tensorflow

🐝Tensorflow Implementation of Spatial Transformer Networks
Python
290
star
4

code-against-climate-change

🌏 A curated list of tech projects against climate change - hoping to inspire disruptive technological climate action!
192
star
5

deep-autonomous-driving-papers

🚘 A curated list of papers of deep learning in autonomous driving papers
94
star
6

green-ai

🌱 The Green AI Standard aims to develop a standard and raise awareness for best environmental practices in AI research and development
77
star
7

awesome-data-valuation

💱 A curated list of data valuation (DV) to design your next data marketplace
63
star
8

exelixis

Interactive and easy-to-use phylogenetic tree viewer for the web
JavaScript
14
star
9

biojs-io-newick

Newick Parser in JS - parses newick strings into JSON and JSON into newick
JavaScript
14
star
10

pytorch-without-a-phd

PyTorch version of TensorFlow without a PhD
Jupyter Notebook
9
star
11

pytorch-neural-search-optimizer

PyTorch implementation of Neural Optimizer Search's Optimizer_1
Python
8
star
12

deep-learning-slides

Deep Learning in Action is an event organized by @munichacm - Materials 🎥 📘!
7
star
13

prediction-market-tutorial

How to build your own predictive market
JavaScript
6
star
14

attentive-siamese-cnn

Siamese CNN with STN to learn replicate feature maps
Jupyter Notebook
5
star
15

biojs-rest-ensembl

REST API for the ensembl website
JavaScript
4
star
16

nvidia-deep-learning-tutorial

IPython Notebooks from the NVIDIA tutorial at Harvard ComputeFest (Torch, Caffe, Theano)
Jupyter Notebook
3
star
17

tsne-algorithm

This is a fork of the tSNEJS library for npm
JavaScript
2
star
18

LxMLS-labs-solution

My solutions for the Machine Learning Summer School in Lisbon 2015
Python
1
star
19

biojs-workshopper

an automated workshopper build for easier following workshop instructions in biojs [under construction]
CSS
1
star
20

GPU-Programming-TUM

This is a repository for the GPU Programming Lab Course at TUM
Cuda
1
star
21

shape-analysis

MATLAB programming solutions for the TUM lecture "Analysis of Three-Dimensional Shapes"
MATLAB
1
star
22

CellProfiler-REST

A REST API for CellProfiler Classifier
Python
1
star
23

3mk-script

A script in tex for the KIT lecture "multilingual human computer interaction"
TeX
1
star
24

BBBC021_cropped_data

Cropped cell dataset from BBBC021, sorted after compound treatment and concentration, DNA stain only, Filename is ObjectKey
Jupyter Notebook
1
star
25

biojs-vis-tsne

Easy to use t-SNE Visualisation for the web
JavaScript
1
star
26

deep-learning-notes

deep learning for the visually impaired
1
star
27

selfdriving

Put HAL9000 into the car!
Python
1
star
28

conv-vae

Convolutional Variational Autoencoder for EC2 GPU instance test
Python
1
star
29

data

data from experiments
Python
1
star
30

dmc

1. Place Data Mining Cup src for the TUM Lecture "Business Analytics".
R
1
star
31

notes

📓 My personal notebook where I collect useful links and readings for all kinds of topics
1
star
32

homebrew-rumble

Ruby
1
star