Summary of Privacy-preserving Distributed Learning For Residential Short-term Load Forecasting, by Yi Dong et al.
Privacy-Preserving Distributed Learning for Residential Short-Term Load Forecasting
by Yi Dong, Yingjie Wang, Mariana Gama, Mustafa A. Mustafa, Geert Deconinck, Xiaowei Huang
First submitted to arxiv on: 2 Feb 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR); Distributed, Parallel, and Cluster Computing (cs.DC); Multiagent Systems (cs.MA); Systems and Control (eess.SY)
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 abstract presents a novel approach to secure and robust load forecasting in power systems, addressing concerns about data privacy and vulnerabilities to emerging attack techniques. The authors introduce Secure-Aggregation (SecAgg) algorithm, which leverages multiparty computation cryptographic techniques to mitigate gradient leakage risk. However, SecAgg requires additional sub-center servers, increasing computational complexity and reducing system robustness. To address these challenges, the authors propose a Markovian Switching-based distributed training framework (DMS), demonstrating strong robustness against poisoning attacks through rigorous theoretical analysis. Case studies employing real-world power system load data validate the efficacy of the proposed algorithm, achieving significant communication complexity reduction while maintaining accuracy levels comparable to traditional federated learning methods. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary A team of researchers is working on a new way to predict how much energy people will use at home. This matters because some people might not want others to know when they’re home or away. To keep their information private, the team uses a method called federated learning. However, this method has some weaknesses that could be exploited by attackers. The researchers developed two new methods to make it harder for attackers to get information: Secure-Aggregation and Markovian Switching-based distributed training framework (DMS). DMS is particularly good at stopping attacks that try to poison the system with false information. The team tested their ideas using real-world data and found that they work well. |
Keywords
* Artificial intelligence * Federated learning
