Summary of Multi-objective Differentiable Neural Architecture Search, by Rhea Sanjay Sukthanker et al.
Multi-objective Differentiable Neural Architecture Search
by Rhea Sanjay Sukthanker, Arber Zela, Benedikt Staffler, Samuel Dooley, Josif Grabocka, Frank Hutter
First submitted to arxiv on: 28 Feb 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (stat.ML)
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 The proposed neural architecture search (NAS) algorithm encodes user preferences to trade-off performance and hardware metrics in multi-objective optimization (MOO). This novel approach yields representative and diverse architectures across multiple devices with a single search run. The hypernetwork parameterizes the joint architectural distribution, enabling zero-shot transferability to new devices. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper presents an innovative NAS algorithm that balances performance and hardware metrics for neural architecture search. It uses a hypernetwork to condition on hardware features and preference vectors, allowing for transferability to new devices without additional costs. The method outperforms existing MOO NAS approaches in various search spaces and datasets. |
Keywords
* Artificial intelligence * Optimization * Transferability * Zero shot
