Filters








2 Hits in 0.85 sec

MoGA: Searching Beyond MobileNetV3 [article]

Xiangxiang Chu, Bo Zhang, Ruijun Xu
2020 arXiv   pre-print
With the latest MobileNetV3, neural architecture search again claimed its supremacy in network design.  ...  With 200x fewer GPU days than MnasNet, we obtain a series of models that outperform MobileNetV3 under the similar latency constraints, i.e., MoGA-A achieves 75.9% top-1 accuracy on ImageNet, MoGA-B meets  ...  Finally, we present our searched architectures that outperform MobileNetV3. MoGA-A that achieves 75.9% top-1 accuracy on ImageNet, MoGA-B 75.5% and MoGA-C 75.3%.  ... 
arXiv:1908.01314v4 fatcat:jb2qbtfsgnfczio3po4psyvu2u

Finding Storage- and Compute-Efficient Convolutional Neural Networks

Daniel Becking, Simon Wiedemann, Klaus-Robert Müller
2020 Zenodo  
-Large [1] 75.2 5.4M 219M RL & NA MnasNet-A1 [2] 75.2 3.9M 312M RL MoGA-C [36] 75.3 5.4M 221M EA MnasNet-A2 [2] 75.6 4.8M 340M RL MoGA-A [36] 75.9 5.1M 304M EA EfficientNet-B0 [  ...  This may prevent to find novel architectures that go beyond the current human knowledge.  ... 
doi:10.5281/zenodo.5501151 fatcat:zjh4kngadrgtdgzniphqrvfndq