Summary of Towards Active Participant Centric Vertical Federated Learning: Some Representations May Be All You Need, by Jon Irureta et al.
Towards Active Participant Centric Vertical Federated Learning: Some Representations May Be All You Need
by Jon Irureta, Jon Imaz, Aizea Lojo, Javier Fernandez-Marques, Marco González, Iñigo Perona
First submitted to arxiv on: 23 Oct 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
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 A novel approach to Vertical Federated Learning (VFL) called Active Participant Centric VFL (APC-VFL) is introduced, exceling in scenarios with partially aligned data among participants. This method requires only a single communication step with the active participant, made possible through local and unsupervised representation learning at each participant followed by knowledge distillation in the active participant. APC-VFL consistently outperforms existing VFL methods like SplitNN or VFedTrans across three popular VFL datasets, achieving better F1 scores, accuracy, and communication costs as the ratio of aligned data is reduced. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary A new way to learn from different sources called Vertical Federated Learning (VFL) makes it easier for devices to work together. This new method, called Active Participant Centric VFL (APC-VFL), is better when some data is shared among these devices. It only needs one message sent between the devices and does a good job of learning from this shared data. APC-VFL works well on three big datasets and is faster than other methods. |
Keywords
» Artificial intelligence » Federated learning » Knowledge distillation » Representation learning » Unsupervised
