Summary of Combining Neural Architecture Search and Automatic Code Optimization: a Survey, by Inas Bachiri et al.
Combining Neural Architecture Search and Automatic Code Optimization: A Survey
by Inas Bachiri, Hadjer Benmeziane, Smail Niar, Riyadh Baghdadi, Hamza Ouarnoughi, Abdelkrime Aries
First submitted to arxiv on: 7 Aug 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Programming Languages (cs.PL)
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 This paper explores how to accelerate complex Deep Learning models for efficient execution on resource-constrained devices, such as mobile phones or embedded systems. The authors focus on two techniques: Hardware-aware Neural Architecture Search (HW-NAS) and Automatic Code Optimization (ACO). HW-NAS automatically designs neural networks that are optimized for specific hardware, while ACO searches for the best compiler optimizations to apply to these networks. The paper presents a survey of recent works that combine these two techniques into a single framework, demonstrating their sub-optimality when performed independently. By integrating them, the authors propose a joint optimization process called Hardware Aware-Neural Architecture and Compiler Optimizations co-Search (NACOS). |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper is about making Deep Learning models run faster on devices with limited resources. It talks about two ways to do this: HW-NAS and ACO. HW-NAS helps design neural networks that are good for specific hardware, while ACO finds the best way to compile these networks for efficient execution. The authors look at how combining these techniques can make things better. They show that doing them separately isn’t as good as doing them together. |
Keywords
» Artificial intelligence » Deep learning » Optimization
