Incentivizing Massive Unknown Workers for Budget-Limited Crowdsensing: From Off-Line and On-Line Perspectives

TL;DR

This work addresses budget-limited crowdsensing with massive unknown workers by introducing offline and online Context-Aware CMAB-based Incentive (CACI) mechanisms that learn worker quality through partitioned context hypercubes rather than per-worker arms. By combining exploration of context hypercubes with UCB-based exploitation, CACI achieves provable regret bounds of $oxed{ ilde{O}ig(B^{(2oldsymbol{ extalpha}+M)/(3oldsymbol{ extalpha}+M)}ig)}$ for offline and an $O(B)$-type bound for online settings, while guaranteeing truthfulness and individual rationality. The offline mechanism uses a two-phase approach over context partitions, and the online variant adapts the learning to dynamic worker arrivals with context-based inference of sensing abilities. Extensive experiments on synthetic and real datasets demonstrate that CACIs outperform baselines in cumulative reward and exhibit robust performance under tight budgets, highlighting practical impact for scalable, incentive-compatible crowdsensing systems.

Abstract

How to incentivize strategic workers using limited budget is a very fundamental problem for crowdsensing systems; nevertheless, since the sensing abilities of the workers may not always be known as prior knowledge due to the diversities of their sensor devices and behaviors, it is difficult to properly select and pay the unknown workers. Although the uncertainties of the workers can be addressed by the standard Combinatorial Multi-Armed Bandit (CMAB) framework in existing proposals through a trade-off between exploration and exploitation, we may not have sufficient budget to enable the trade-off among the individual workers, especially when the number of the workers is huge while the budget is limited. Moreover, the standard CMAB usually assumes the workers always stay in the system, whereas the workers may join in or depart from the system over time, such that what we have learnt for an individual worker cannot be applied after the worker leaves. To address the above challenging issues, in this paper, we first propose an off-line Context-Aware CMAB-based Incentive (CACI) mechanism. We innovate in leveraging the exploration-exploitation trade-off in an elaborately partitioned context space instead of the individual workers, to effectively incentivize the massive unknown workers with a very limited budget. We also extend the above basic idea to the on-line setting where unknown workers may join in or depart from the systems dynamically, and propose an on-line version of the CACI mechanism. We perform rigorous theoretical analysis to reveal the upper bounds on the regrets of our CACI mechanisms and to prove their truthfulness and individual rationality, respectively. Extensive experiments on both synthetic and real datasets are also conducted to verify the efficacy of our mechanisms.

Figures (10)

• Figure 1: Comparison results under different budgets on synthetic dataset.
• Figure 2: Comparison results with different numbers of workers on synthetic dataset.
• Figure 3: Truthfulness and individual rationality.
• Figure 4: Comparison results under different budgets on vehicular trajectory dataset.
• Figure 5: Comparison results with different numbers of workers on vehicular trajectory dataset.

Theorems & Definitions (28)

• Definition 1
• Definition 2
• Proposition 1
• Proposition 2
• Lemma 1
• proof
• Theorem 1
• proof
• Proposition 3
• Proposition 4