HyT-NAS: Hybrid Transformers Neural Architecture Search for Edge Devices
This work addresses the challenge of implementing attention-based models on tiny devices, offering a domain-specific incremental improvement.
The paper tackled the problem of deploying vision transformers on resource-constrained edge devices by proposing HyT-NAS, an efficient hardware-aware neural architecture search method, resulting in a 6.3% accuracy improvement over MLPerf MobileNetV1 with 3.5x fewer parameters on Visual Wake Words.
Vision Transformers have enabled recent attention-based Deep Learning (DL) architectures to achieve remarkable results in Computer Vision (CV) tasks. However, due to the extensive computational resources required, these architectures are rarely implemented on resource-constrained platforms. Current research investigates hybrid handcrafted convolution-based and attention-based models for CV tasks such as image classification and object detection. In this paper, we propose HyT-NAS, an efficient Hardware-aware Neural Architecture Search (HW-NAS) including hybrid architectures targeting vision tasks on tiny devices. HyT-NAS improves state-of-the-art HW-NAS by enriching the search space and enhancing the search strategy as well as the performance predictors. Our experiments show that HyT-NAS achieves a similar hypervolume with less than ~5x training evaluations. Our resulting architecture outperforms MLPerf MobileNetV1 by 6.3% accuracy improvement with 3.5x less number of parameters on Visual Wake Words.