NeuralScale: Efficient Scaling of Neurons for Resource-Constrained Deep Neural Networks

By Eugene Lee, Chen-Yi Lee. 2020

15 citations recorded by the source corpus.

Deciding the amount of neurons during the design of a deep neural network to maximize performance is not intuitive. In this work, we attempt to search for the neuron (filter) configuration of a fixed network architecture that maximizes accuracy. Using iterative pruning methods as a proxy, we parametrize the change of the neuron (filter) number of each layer with respect to the change in parameters, allowing us to efficiently scale an architecture across arbitrary sizes. We also introduce architecture descent which iteratively refines the parametrized function used for model scaling. The combination of both proposed methods is coined as NeuralScale. To prove the efficiency of NeuralScale in terms of parameters, we show empirical simulations on VGG11, MobileNetV2 and ResNet18 using CIFAR10, CIFAR100 and TinyImageNet as benchmark datasets. Our results show an increase in accuracy of 3.04%, 8.56% and 3.41% for VGG11, MobileNetV2 and ResNet18 on CIFAR10, CIFAR100 and TinyImageNet respectively under a parameter-constrained setting (output neurons (filters) of default configuration with scaling factor of 0.25).

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Chen-Yi Lee
2. Cheng-Han Huang
3. Eugene Lee

In Computer Vision and Pattern Recognition

1. Chen-Yi Lee
2. Cheng-Han Huang
3. Eugene Lee

Citations in the corpus

• Few-Shot and Continual Learning with Attentive Independent Mechanisms (2021, background): “Tremendous effort is placed into neural architecture search (NAS) [28, 54, 49, 39, 32] with the hypothesis that improvements on a structural level introduce inductive bias that improves the generalizability of a neural network.”