Summary of Mp-sl: Multihop Parallel Split Learning, by Joana Tirana et al.
MP-SL: Multihop Parallel Split Learning
by Joana Tirana, Spyros Lalis, Dimitris Chatzopoulos
First submitted to arxiv on: 31 Jan 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Distributed, Parallel, and Cluster Computing (cs.DC)
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 introduces Multihop Parallel Split Learning (MP-SL), a Federated Learning framework designed for collaborative and distributed Machine Learning model training on resource-constrained devices. MP-SL addresses the challenges of heterogeneous device participation by splitting models into parts, utilizing multiple compute nodes in a pipelined manner to alleviate memory demands per node. The authors demonstrate the efficiency of MP-SL in scenarios involving more cost-effective compute nodes, outperforming horizontally scaled one-hop Parallel SL setups. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary In this paper, scientists create a new way for devices with limited resources to work together and train machine learning models. They call it Multihop Parallel Split Learning (MP-SL). This helps smaller devices like mobile phones and IoT devices participate in training models alongside more powerful computers. MP-SL splits the model into parts and uses multiple computers in a line to process the information, reducing memory needs for each computer. The researchers show that this method works well, especially when using cheaper but still capable computers. |
Keywords
* Artificial intelligence * Federated learning * Machine learning
