Efficient Neural Architecture Search (ENAS) in PyTorch

PyTorch implementation of Efficient Neural Architecture Search via Parameters Sharing.

ENAS reduce the computational requirement (GPU-hours) of Neural Architecture Search (NAS) by 1000x via parameter sharing between models that are subgraphs within a large computational graph. SOTA on Penn Treebank language modeling.

**[Caveat] Use official code from the authors: link**

Prerequisites

• Python 3.6+
• PyTorch==0.3.1
• tqdm, scipy, imageio, graphviz, tensorboardX

Usage

Install prerequisites with:

conda install graphviz
pip install -r requirements.txt

To train ENAS to discover a recurrent cell for RNN:

python main.py --network_type rnn --dataset ptb --controller_optim adam --controller_lr 0.00035 \
               --shared_optim sgd --shared_lr 20.0 --entropy_coeff 0.0001

python main.py --network_type rnn --dataset wikitext

To train ENAS to discover CNN architecture (in progress):

python main.py --network_type cnn --dataset cifar --controller_optim momentum --controller_lr_cosine=True \
               --controller_lr_max 0.05 --controller_lr_min 0.0001 --entropy_coeff 0.1

or you can use your own dataset by placing images like:

data
├── YOUR_TEXT_DATASET
│   ├── test.txt
│   ├── train.txt
│   └── valid.txt
├── YOUR_IMAGE_DATASET
│   ├── test
│   │   ├── xxx.jpg (name doesn't matter)
│   │   ├── yyy.jpg (name doesn't matter)
│   │   └── ...
│   ├── train
│   │   ├── xxx.jpg
│   │   └── ...
│   └── valid
│       ├── xxx.jpg
│       └── ...
├── image.py
└── text.py

To generate gif image of generated samples:

python generate_gif.py --model_name=ptb_2018-02-15_11-20-02 --output=sample.gif

More configurations can be found here.

Results

Efficient Neural Architecture Search (ENAS) is composed of two sets of learnable parameters, controller LSTM θ and the shared parameters ω. These two parameters are alternatively trained and only trained controller is used to derive novel architectures.

1. Discovering Recurrent Cells

Controller LSTM decide 1) what activation function to use and 2) which previous node to connect.

The RNN cell ENAS discovered for Penn Treebank and WikiText-2 dataset:

Best discovered ENAS cell for Penn Treebank at epoch 27:

You can see the details of training (e.g. reward, entropy, loss) with:

tensorboard --logdir=logs --port=6006

2. Discovering Convolutional Neural Networks

Controller LSTM samples 1) what computation operation to use and 2) which previous node to connect.

The CNN network ENAS discovered for CIFAR-10 dataset:

(in progress)

3. Designing Convolutional Cells

(in progress)

Reference

• Neural Architecture Search with Reinforcement Learning
• Neural Optimizer Search with Reinforcement Learning

Author

Taehoon Kim / @carpedm20