Summary of Iotco2: Assessing the End-to-end Carbon Footprint Of Internet-of-things-enabled Deep Learning, by Fan Chen et al.
IoTCO2: Assessing the End-To-End Carbon Footprint of Internet-of-Things-Enabled Deep Learning
by Fan Chen, Shahzeen Attari, Gayle Buck, Lei Jiang
First submitted to arxiv on: 16 Mar 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
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 proposes an end-to-end tool called for estimating the carbon footprint associated with deep learning (DL) models deployed on Internet of Things (IoT) devices. The tool aims to improve privacy and ensure quality-of-service (QoS) by accurately predicting both operational and embodied aspects of DL on IoT, covering quantized DL models and emerging neural processing units (NPUs). Existing predictors often overlook these factors, creating a gap in accurate carbon footprint modeling tools for IoT-enabled DL. achieves deviations as low as 5% for operational and 3.23% for embodied carbon footprints compared to actual measurements across various DL models, with practical applications showcased through multiple user case studies. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper creates a tool called that helps measure the environmental impact of using deep learning on devices like smart home devices or wearables. Right now, we don’t have good ways to estimate how much energy and resources these devices use when processing data. This tool tries to fill this gap by being more accurate in its predictions, which can help us make better decisions about where to deploy our models and reduce the negative effects on the environment. |
Keywords
* Artificial intelligence * Deep learning
