Google MorphNet makes your neural net faster and smaller
Google MorphNet makes your neural net faster and smaller
MorphNet is a Google Model to Build Faster and Smaller Neural Networks
Towards AI article https://pub.towardsai.net/morphnet-is-a-google-model-to-build-faster-and-smaller-neural-networks-f890276da456
Fine-Grained Stochastic Architecture Search
arXiv paper abstract https://arxiv.org/abs/2006.09581
arXiv PDf paper https://arxiv.org/pdf/2006.09581.pdf
State-of-the-art deep networks are often too large to deploy on mobile devices and embedded systems.
... introduce Fine-Grained Stochastic Architecture Search (FiGS)
... FiGS simultaneously selects and modifies operators in the search space by applying a structured sparse regularization penalty based on the Logistic-Sigmoid distribution.
... producing state-of-the-art parameter-efficient models on ImageNet (e.g., 75.4% top-1 with 2.6M params).
... we achieve significantly higher mAP on COCO (e.g., 25.8 with 3.0M params) than MobileNetV3 and MnasNet.
Please like and share this post if you enjoyed it using the buttons at the bottom!
Stay up to date. Subscribe to my posts https://morrislee1234.wixsite.com/website/contact
Web site with my other posts by category https://morrislee1234.wixsite.com/website
Comments