Contact: dkelesis@iit.demokritos.gr
Public goods provision in social networks involves designing mechanisms to efficiently allocate resources for public goods in a way that maximizes social welfare. Traditional methods include game-theoretic approaches and incentive mechanisms. Recent research has explored the use of graph-based methods to model social networks and design efficient allocation mechanisms.
This thesis will focus on the development of graph neural network (GNN)-based approaches for public goods provision. Research questions include: How can GNNs be used to model the dynamics of public goods provision in social networks? What are the benefits of using GNN-based methods in terms of efficiency and fairness? Can GNNs help design more robust and adaptable allocation mechanisms for dynamic social networks?
Relevant literature:
[1] https://arxiv.org/pdf/2106.09761
[2] https://iopscience.iop.org/article/10.1088/2632-2153/ac4d12
[3] https://ieeexplore.ieee.org/document/9322537
[4] https://sands.kaust.edu.sa/classes/CS294E/F21/papers/graf.pdf