Deep learning

Channel (broadcasting)

Meta learning (computer science)

Residual neural network

Computer network

Agronomy

Biology

Pruning

Epistemology

Machine learning

Philosophy

Artificial intelligence

Management

Economics

Computer science

Task (project management)

Simple (philosophy)

Deep neural networks

Artificial neural network

In this paper, we propose a novel meta learning approach for automatic channel pruning of very deep neural networks. We first train a PruningNet, a kind of meta network, which is able to generate weight parameters for any pruned structure given the target network. We use a simple stochastic structure sampling method for training the PruningNet. Then, we apply an evolutionary procedure to search for good-performing pruned networks. The search is highly efficient because the weights are directly generated by the trained PruningNet and we do not need any finetuning at search time. With a single PruningNet trained for the target network, we can search for various Pruned Networks under different constraints with little human participation. Compared to the state-of-the-art pruning methods, we have demonstrated superior performances on MobileNet V1/V2 and ResNet. Codes are available on https://github.com/liuzechun/MetaPruning.

MetaPruning: Meta Learning for Automatic Neural Network Channel Pruning