# RADARS: Memory Efficient Reinforcement Learning Aided Differentiable Neural Architecture Search

arXiv: https://arxiv.org/abs/2109.05691 · HTML: https://mlsanity.com/paper/2109.05691/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 20 references that are in the map.
- First posted on arXiv 13 Sep 2021.

## Builds on (most cited first, 18 of 18)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [MobileNetV2: Inverted Residuals and Linear Bottlenecks](https://mlsanity.com/paper/1801.04381/) (2018)
- [DARTS: Differentiable Architecture Search](https://mlsanity.com/paper/1806.09055/) (2018)
- [Neural Architecture Search with Reinforcement Learning](https://mlsanity.com/paper/1611.01578/) (2016)
- [Learning Transferable Architectures for Scalable Image Recognition](https://mlsanity.com/paper/1707.07012/) (2017)
- [ProxylessNAS: Direct Neural Architecture Search on Target Task and Hardware](https://mlsanity.com/paper/1812.00332/) (2018)
- [Progressive Neural Architecture Search](https://mlsanity.com/paper/1712.00559/) (2017)
- [CondenseNet: An Efficient DenseNet using Learned Group Convolutions](https://mlsanity.com/paper/1711.09224/) (2017)
- [One-Shot Neural Architecture Search via Self-Evaluated Template Network](https://mlsanity.com/paper/1910.05733/) (2019)
- [Hardware/Software Co-Exploration of Neural Architectures](https://mlsanity.com/paper/1907.04650/) (2019)
- [Few-shot Neural Architecture Search](https://mlsanity.com/paper/2006.06863/) (2020)
- [Co-Exploration of Neural Architectures and Heterogeneous ASIC Accelerator Designs Targeting Multiple Tasks](https://mlsanity.com/paper/2002.04116/) (2020)
- [Accuracy vs. Efficiency: Achieving Both through FPGA-Implementation Aware Neural Architecture Search](https://mlsanity.com/paper/1901.11211/) (2019)
- [EDD: Efficient Differentiable DNN Architecture and Implementation Co-search for Embedded AI Solutions](https://mlsanity.com/paper/2005.02563/) (2020)
- [Device-Circuit-Architecture Co-Exploration for Computing-in-Memory Neural Accelerators](https://mlsanity.com/paper/1911.00139/) (2019)
- [UNAS: Differentiable Architecture Search Meets Reinforcement Learning](https://mlsanity.com/paper/1912.07651/) (2019)
- [On Neural Architecture Search for Resource-Constrained Hardware Platforms](https://mlsanity.com/paper/1911.00105/) (2019)
- [Uncertainty Modeling of Emerging Device-based Computing-in-Memory Neural Accelerators with Application to Neural Architecture Search](https://mlsanity.com/paper/2107.06871/) (2021)

## Built on by (most cited first, 8 of 8)

- [SWIM: Selective Write-Verify for Computing-in-Memory Neural Accelerators](https://mlsanity.com/paper/2202.08395/) (2022)
- [Computing-In-Memory Neural Network Accelerators for Safety-Critical Systems: Can Small Device Variations Be Disastrous?](https://mlsanity.com/paper/2207.07626/) (2022)
- [RRNet: Towards ReLU-Reduced Neural Network for Two-party Computation Based Private Inference](https://mlsanity.com/paper/2302.02292/) (2023)
- [Improving Realistic Worst-Case Performance of NVCiM DNN Accelerators through Training with Right-Censored Gaussian Noise](https://mlsanity.com/paper/2307.15853/) (2023)
- [Efficient Automation of Neural Network Design: A Survey on Differentiable Neural Architecture Search](https://mlsanity.com/paper/2304.05405/) (2023)
- [Compute-in-Memory based Neural Network Accelerators for Safety-Critical Systems: Worst-Case Scenarios and Protections](https://mlsanity.com/paper/2312.06137/) (2023)
- [U-SWIM: Universal Selective Write-Verify for Computing-in-Memory Neural Accelerators](https://mlsanity.com/paper/2401.05357/) (2023)
- [An Integrated Approach to Neural Architecture Search for Deep Q-Networks](https://mlsanity.com/paper/2510.19872/) (2025)

## Abstract (arXiv)

Differentiable neural architecture search (DNAS) is known for its capacity in the automatic generation of superior neural networks. However, DNAS based methods suffer from memory usage explosion when the search space expands, which may prevent them from running successfully on even advanced GPU platforms. On the other hand, reinforcement learning (RL) based methods, while being memory efficient, are extremely time-consuming. Combining the advantages of both types of methods, this paper presents RADARS, a scalable RL-aided DNAS framework that can explore large search spaces in a fast and memory-efficient manner. RADARS iteratively applies RL to prune undesired architecture candidates and identifies a promising subspace to carry out DNAS. Experiments using a workstation with 12 GB GPU memory show that on CIFAR-10 and ImageNet datasets, RADARS can achieve up to 3.41% higher accuracy with 2.5X search time reduction compared with a state-of-the-art RL-based method, while the two DNAS baselines cannot complete due to excessive memory usage or search time. To the best of the authors' knowledge, this is the first DNAS framework that can handle large search spaces with bounded memory usage.