Summary of Anatomizing Deep Learning Inference in Web Browsers, by Qipeng Wang et al.
Anatomizing Deep Learning Inference in Web Browsers
by Qipeng Wang, Shiqi Jiang, Zhenpeng Chen, Xu Cao, Yuanchun Li, Aoyu Li, Yun Ma, Ting Cao, Xuanzhe Liu
First submitted to arxiv on: 8 Feb 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Performance (cs.PF)
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 In this paper, researchers investigate the performance and quality of experience (QoE) of in-browser deep learning (DL) inference. They propose new metrics to measure responsiveness, smoothness, and accuracy, and conduct an extensive analysis across 9 DL models, 50 PC devices, and 20 mobile devices. The results show a significant latency gap between in-browser and native inference on both CPU and GPU, as well as memory demands exceeding the size of the DL models themselves. This research provides valuable insights into the challenges and limitations of in-browser DL inference, highlighting its impact on user QoE. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary In-browser deep learning is a way for websites to use powerful AI models directly within your web browser. Researchers wanted to know how well this works and if it’s good enough. They looked at 9 different AI models, many types of devices, and measured how fast and smooth the AI was. Surprisingly, they found that in-browser AI is much slower than using a computer or phone specifically designed for AI tasks. This can make websites run slowly and be less fun to use. |
Keywords
* Artificial intelligence * Deep learning * Inference
