North carolina state University researchers have developed a new framework for building deep neural networks via grammar-guided network generators. In experimental testing, the new networks - called AOGNets -- have outperformed existing state-of-the-art frameworks, including the widely used ResNet and DenseNet systems, in visual recognition tasks.

To evaluate new network architectures for deep learning in visual recognition, they are the golden testbeds," Wu says, an assistant professor of electrical and computer engineering at NC State ".AOGNets are developed under a principled grammar framework and obtain significant improvement in both ImageNet and MS-COCO, thus showing potentially broad and deep impacts for representation learning in numerous practical applications.

Those tests are relevant because image classification is one of the core basic tasks in visual recognition, and ImageNet is the standard large-scale classification benchmark. Similarly, object detection and segmentation are two core high-level vision tasks, and MS-COCO is one of the most widely used benchmarks.

some facts about new framework(AOGNets) for Neural network - 

  The new framework uses a compositional grammar approach to system architecture that draws on 
  best practices from previous network systems to more effectively extract useful information from 
  raw  data.

  hierarchical and compositional grammar gave us a simple, elegant way to unify the approaches
  taken by previous system architectures, and to our best knowledge, it is the first work that makes 
  use of grammar for network generation," Wu says.

  AOGNets obtained significantly better performance than all of the state-of-the-art networks under
  fair comparisons, including ResNets, DenseNets, ResNeXts and DualPathNets, Wu says. AOGNets 
  also obtained the best model interpretability score using the network dissection metric in ImageNet.
  AOGNets further show great potential in adversarial defense and platform-agnostic deployment 
  (mobile vs cloud).

  The researchers also tested the performance of AOGNets in object detection and instance semantic
  segmentation, on the Microsoft COCO benchmark, using the vanilla Mask R-CNN system.

   AOGNets obtained better results than the ResNet and ResNeXt backbones with smaller model sizes
  and similar or slightly better inference time,  Wu says. "The results show the effectiveness of 
  AOGNets earning better features in object detection and segmentation tasks.


you can go through paper-