Neural Epitome Search for Architecture-Agnostic Network Compression
arXiv:1907.05642
Abstract
The recent WSNet [1] is a new model compression method through sampling filterweights from a compact set and has demonstrated to be effective for 1D convolutionneural networks (CNNs). However, the weights sampling strategy of WSNet ishandcrafted and fixed which may severely limit the expression ability of the resultedCNNs and weaken its compression ability. In this work, we present a novel auto-sampling method that is applicable to both 1D and 2D CNNs with significantperformance improvement over WSNet. Specifically, our proposed auto-samplingmethod learns the sampling rules end-to-end instead of being independent of thenetwork architecture design. With such differentiable weight sampling rule learning,the sampling stride and channel selection from the compact set are optimized toachieve better trade-off between model compression rate and performance. Wedemonstrate that at the same compression ratio, our method outperforms WSNetby6.5% on 1D convolution. Moreover, on ImageNet, our method outperformsMobileNetV2 full model by1.47%in classification accuracy with25%FLOPsreduction. With the same backbone architecture as baseline models, our methodeven outperforms some neural architecture search (NAS) based methods such asAMC [2] and MNasNet [3].
References in corpus (11)
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- Distilling the Knowledge in a Neural Network
- MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
- Quantized Neural Networks: Training Neural Networks with Low Precision Weights and Activations
- DARTS: Differentiable Architecture Search
- AMC: AutoML for Model Compression and Acceleration on Mobile Devices
- Group Equivariant Convolutional Networks
- Speeding up Convolutional Neural Networks with Low Rank Expansions
- Memory Bounded Deep Convolutional Networks
- ASAP: Architecture Search, Anneal and Prune
- ChannelNets: Compact and Efficient Convolutional Neural Networks via Channel-Wise Convolutions