Summary of Growing Efficient Accurate and Robust Neural Networks on the Edge, by Vignesh Sundaresha and Naresh Shanbhag
Growing Efficient Accurate and Robust Neural Networks on the Edge
by Vignesh Sundaresha, Naresh Shanbhag
First submitted to arxiv on: 10 Oct 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Computer Vision and Pattern Recognition (cs.CV)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary Medium Difficulty Summary: The paper presents GEARnn, a novel approach for growing and training deep learning models on resource-constrained edge devices without relying on cloud computing. Traditional methods rely on cloud-based training and compression, which incurs high energy and latency costs while raising privacy concerns. GEARnn employs One-Shot Growth (OSG) to grow a network satisfying memory constraints using clean data and Efficient Robust Augmentation (ERA) for robustness. The proposed method is demonstrated on a NVIDIA Jetson Xavier NX, showing the trade-offs between accuracy, robustness, model size, energy consumption, and training time. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Low Difficulty Summary: This paper talks about how to make artificial intelligence work better on devices like smartphones or smart home devices that don’t have as much power. Right now, these devices need to send their data to the cloud for processing, which takes a lot of energy and time. The researchers propose a new way called GEARnn that lets these devices train their own AI models without needing to go to the cloud. They show how this works on a special device and explain the benefits of faster and more efficient AI. |
Keywords
* Artificial intelligence * Deep learning * One shot
