# DONNAv2 -- Lightweight Neural Architecture Search for Vision tasks

arXiv: https://arxiv.org/abs/2309.14670 · HTML: https://mlsanity.com/paper/2309.14670/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 26 of its 35 references that are in the map.
- First posted on arXiv 26 Sep 2023.

## Builds on (most cited first, 26 of 26)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks](https://mlsanity.com/paper/1905.11946/) (2019)
- [DARTS: Differentiable Architecture Search](https://mlsanity.com/paper/1806.09055/) (2018)
- [Similarity of Neural Network Representations Revisited](https://mlsanity.com/paper/1905.00414/) (2019)
- [Enhanced Deep Residual Networks for Single Image Super-Resolution](https://mlsanity.com/paper/1707.02921/) (2017)
- [Neural Architecture Search with Reinforcement Learning](https://mlsanity.com/paper/1611.01578/) (2016)
- [BDD100K: A Diverse Driving Dataset for Heterogeneous Multitask Learning](https://mlsanity.com/paper/1805.04687/) (2018)
- [Learning Transferable Architectures for Scalable Image Recognition](https://mlsanity.com/paper/1707.07012/) (2017)
- [EfficientDet: Scalable and Efficient Object Detection](https://mlsanity.com/paper/1911.09070/) (2019)
- [MnasNet: Platform-Aware Neural Architecture Search for Mobile](https://mlsanity.com/paper/1807.11626/) (2018)
- [Once-for-All: Train One Network and Specialize it for Efficient Deployment](https://mlsanity.com/paper/1908.09791/) (2019)
- [Simple Baselines for Image Restoration](https://mlsanity.com/paper/2204.04676/) (2022)
- [ProxylessNAS: Direct Neural Architecture Search on Target Task and Hardware](https://mlsanity.com/paper/1812.00332/) (2018)
- [FBNet: Hardware-Aware Efficient ConvNet Design via Differentiable Neural Architecture Search](https://mlsanity.com/paper/1812.03443/) (2018)
- [Zero-Cost Proxies for Lightweight NAS](https://mlsanity.com/paper/2101.08134/) (2021)
- [FBNetV2: Differentiable Neural Architecture Search for Spatial and Channel Dimensions](https://mlsanity.com/paper/2004.05565/) (2020)
- [BRP-NAS: Prediction-based NAS using GCNs](https://mlsanity.com/paper/2007.08668/) (2020)
- [Single-Path NAS: Designing Hardware-Efficient ConvNets in less than 4 Hours](https://mlsanity.com/paper/1904.02877/) (2019)
- [YOLOP: You Only Look Once for Panoptic Driving Perception](https://mlsanity.com/paper/2108.11250/) (2021)
- [Blockwisely Supervised Neural Architecture Search with Knowledge Distillation](https://mlsanity.com/paper/1911.13053/) (2019)
- [Distilling Optimal Neural Networks: Rapid Search in Diverse Spaces](https://mlsanity.com/paper/2012.08859/) (2020)
- [Powering One-shot Topological NAS with Stabilized Share-parameter Proxy](https://mlsanity.com/paper/2005.10511/) (2020)
- [Parallel Blockwise Knowledge Distillation for Deep Neural Network Compression](https://mlsanity.com/paper/2012.03096/) (2020)
- [BLOX: Macro Neural Architecture Search Benchmark and Algorithms](https://mlsanity.com/paper/2210.07271/) (2022)
- [LayerNAS: Neural Architecture Search in Polynomial Complexity](https://mlsanity.com/paper/2304.11517/) (2023)

## Abstract (arXiv)

With the growing demand for vision applications and deployment across edge devices, the development of hardware-friendly architectures that maintain performance during device deployment becomes crucial. Neural architecture search (NAS) techniques explore various approaches to discover efficient architectures for diverse learning tasks in a computationally efficient manner. In this paper, we present the next-generation neural architecture design for computationally efficient neural architecture distillation - DONNAv2 . Conventional NAS algorithms rely on a computationally extensive stage where an accuracy predictor is learned to estimate model performance within search space. This building of accuracy predictors helps them predict the performance of models that are not being finetuned. Here, we have developed an elegant approach to eliminate building the accuracy predictor and extend DONNA to a computationally efficient setting. The loss metric of individual blocks forming the network serves as the surrogate performance measure for the sampled models in the NAS search stage. To validate the performance of DONNAv2 we have performed extensive experiments involving a range of diverse vision tasks including classification, object detection, image denoising, super-resolution, and panoptic perception network (YOLOP). The hardware-in-the-loop experiments were carried out using the Samsung Galaxy S10 mobile platform. Notably, DONNAv2 reduces the computational cost of DONNA by 10x for the larger datasets. Furthermore, to improve the quality of NAS search space, DONNAv2 leverages a block knowledge distillation filter to remove blocks with high inference costs.