Sybil-Proof Mechanism for Information Propagation with Budgets

TL;DR

PRDM tackles budget-constrained reward distribution for information propagation on social networks by introducing a two-phase, layered diffusion mechanism. It computes layer-aware contribution weights and then propagates rewards to parents with a diffusion parameter $\beta$, anchored by a virtual sponsor capacity $c_s$; the design aims to maximize propagation under a fixed budget $B$. The authors prove PRDM is IR, IC, and (parallel) Sybil-proof, and is asymptotically budget-balanced, with a $\frac{1}{1-\beta}$-Sybil-proof bound under a majority assumption. They also discuss inherent trade-offs between diffusion incentives and Sybil resistance, and acknowledge limitations such as collusion, suggesting avenues for future work.

Abstract

This paper examines the problem of distributing rewards on social networks to improve the efficiency of crowdsourcing tasks for sponsors. To complete the tasks efficiently, we aim to design reward mechanisms that incentivize early-joining agents to invite more participants to the tasks. Nonetheless, participants could potentially engage in strategic behaviors, e.g., not inviting others to the tasks, misreporting their capacity for the tasks, or creaking fake identities (aka Sybil attacks), to maximize their own rewards. The focus of this study is to address the challenge outlined above by designing effective reward mechanisms. To this end, we propose a novel reward mechanism, called Propagation Reward Distribution Mechanism (PRDM), for the general information propagation model with limited budgets. It is proved that the PRDM can not only incentivize all agents to contribute their full efforts to the tasks and share the task information to all their neighbors in the social networks, but can also prevent them from Sybil attacks.

Figures (5)

• Figure 1: An example of transforming an active network (a) into a layered graph (b).
• Figure 2: An example of PRDM on input $B=100$, $c_s=20$, $\beta=0.2$, each agent has a contribution of $10$. (a) the invitation relationship among the sponsor and each agent. (b) each layer's initial budget $B_k$ and each agent's weight $w_i$ in contribution phase. (c) the transfer of reward during propagation phase and each agent's final reward $r_i$.
• Figure 3: (a) is the case where agent $i$ does not commit Sybil attacks, the black node represents agent $i$, and the white nodes represent real participants that $i$ invites. (b) shows the situation where $i$ conducts fake nodes one layer down in which the dashed node represent all the nodes generated by $i$. (c) is the most general form of a Sybil attacks.
• Figure 4: The strategies agent $1$ may adopt: agent $1$ can transfer $\delta$ ($0 < \delta < 10$) of her contribution to agent $9$ and she can disinvite agent $4$.
• Figure 5: Relationship between agent $1$'s total utility ($r_1+r_9$) when agent $1$ transfers $\delta$ of her contribution to her fake nodes ($0 < \delta < 10$), under both conditions whether she invites agent $4$.

Theorems & Definitions (28)

• Definition 1
• Definition 2
• Definition 3
• Definition 4
• Definition 5
• Definition 6
• Definition 7
• Definition 8
• Definition 9
• Definition 10