[TMP-108] Social AI gossiping

Contact person: Andrea Passarella, (andrea.passarella@iit.cnr.it)

Internal Partners:

1. Consiglio Nazionale delle ricerche (CNR), Andrea Passarella,

andrea.passarella@iit.cnr.it

2. Central European University, János Kertész, kerteszj@ceu.edu

We envision a human-AI ecosystem in which AI-enabled devices act as proxies of humans and try to learn collectively a model in a decentralized way. Each device learns a local model that needs to be combined with the models learned by the other nodes, in order to improve both the local and global knowledge. The challenge of doing so in a fully decentralized AI system entails understanding how to compose models coming from heterogeneous sources and, in case of potentially untrustworthy nodes, decide who can be trusted and why. In this micro-project, we focus on the specific scenario of model “gossiping” for accomplishing a decentralized learning task and we study what models emerge from the combination of local models, where the combination takes into account the social relationships between the humans associated with the AI. We use synthetic graphs to represent social relationships, and large-scale simulation for performance evaluation.

Results Summary

The micro project has developed a modular simulation framework to test decentralised machine learning algorithms on top of large-scale complex social networks. The framework is written in Python, exploiting state-of-the-art libraries such as networkx (to generate network models) and Pytorch (to implement ML models). The simulator is modular, as it accepts networks in the form of datasets as well as synthetic models. Local data are allocated on each node, which trains a local ML model of choice. Communication rounds are implemented, through which local models are aggregated and re-trained based on local data. Benchmarks are included, namely federated learning and centralised learning. Initial simulation results have been derived, to assess the accuracy of decentralised learning (social AI gossiping) on Barabasi-Albert networks, showing that social AI gossiping is able to achieve comparable accuracy with respect to centralised and federated learning versions (which rely on centralised elements, though). The work has been continued in a follow-up micro-project (TMP-034). The simulator developed in this micro-project and in its follow-up (TMP-034) has been used in the publications reported here.

Tangible Outcomes

1. Palmieri, Luigi, Lorenzo Valerio, Chiara Boldrini, and Andrea Passarella. “”The effect of network topologies on fully decentralized learning: a preliminary investigation.”” In Proceedings of the 1st International Workshop on Networked AI Systems, pp. 1-6. 2023. https://dl.acm.org/doi/10.1145/3597062.3597280 
2. Luigi Palmieri, Lorenzo Valerio, Chiara Boldrini, Andrea Passarella, Marco Conti, “”Exploring the Impact of Disrupted Peer-to-Peer Communications on Fully Decentralized Learning in Disaster Scenarios””, 8th International Conference on Information and Communication Technologies for Disaster Management (ICT-DM 2023). https://doi.org/10.1109/ICT-DM58371.2023.10286953   https://arxiv.org/abs/2310.02986 
3. Luigi Palmieri, Chiara Boldrini, Lorenzo Valerio, Andrea Passarella, and Marco Conti. “Impact of Network Topology on the Performance of Decentralized Federated Learning.” Computer Networks 253 (2024). https://www.sciencedirect.com/science/article/pii/S1389128624005139 
4. [arxiv] Valerio, L., Boldrini, C., Passarella, A., Kertész, J., Karsai, M., and Iñiguez, G., “Coordination-free Decentralised Federated Learning on Complex Networks: Overcoming Heterogeneity”, arXiv e-prints, 2023. https://arxiv.org/abs/2312.04504 
5. [arxiv] Palmieri, Luigi, Chiara Boldrini, Lorenzo Valerio, Andrea Passarella, Marco Conti, and János Kertész. “”Robustness of Decentralised Learning to Nodes and Data Disruption.”” arXiv preprint arXiv:2405.02377 (2024). https://arxiv.org/abs/2405.02377 
6. Code: SAIsim, C. Boldrini, L. Valerio, A. Passarella, https://zenodo.org/record/5780042#.Ybi2sX3MLPw