Diffusion Model-based Incentive Mechanism with Prospect Theory for Edge AIGC Services in 6G IoT

TL;DR

This work tackles incentive design for edge AIGC services in 6G-IoT under information asymmetry, where clients must contract with multiple ASPs possessing private AIGC model complexities. It combines Prospect Theory to capture subjective client utilities with a diffusion-based Soft Actor-Critic (SAC) framework to generate optimal contracts under PT, leveraging Generative Diffusion Models for high-dimensional policy optimization. The authors formulate a contract theory problem with IR/IC constraints, derive feasibility conditions and an explicit form for the optimal rewards, and then implement a diffusion-based learning procedure to produce contract designs that maximize the client’s PT-based utility. Numerical results show the proposed diffusion-based approach outperforming traditional DRL methods and demonstrate the effectiveness of PT in capturing client behavior, paving the way for robust, real-time incentive mechanisms in future 6G-IoT-edge AIGC ecosystems.

Abstract

The fusion of the Internet of Things (IoT) with Sixth-Generation (6G) technology has significant potential to revolutionize the IoT landscape. With the ultra-reliable and low-latency communication capabilities of 6G, 6G-IoT networks can transmit high-quality and diverse data to enhance edge learning. Artificial Intelligence-Generated Content (AIGC) harnesses advanced AI algorithms to automatically generate various types of content. The emergence of edge AIGC integrates with edge networks, facilitating real-time provision of customized AIGC services by deploying AIGC models on edge devices. However, the current practice of edge devices as AIGC Service Providers (ASPs) lacks incentives, hindering the sustainable provision of high-quality edge AIGC services amidst information asymmetry. In this paper, we develop a user-centric incentive mechanism framework for edge AIGC services in 6G-IoT networks. Specifically, we first propose a contract theory model for incentivizing ASPs to provide AIGC services to clients. Recognizing the irrationality of clients towards personalized AIGC services, we utilize Prospect Theory (PT) to capture their subjective utility better. Furthermore, we adopt the diffusion-based soft actor-critic algorithm to generate the optimal contract design under PT, outperforming traditional deep reinforcement learning algorithms. Our numerical results demonstrate the effectiveness of the proposed scheme.

Figures (8)

• Figure 1: A user-centric incentive mechanism framework for edge AIGC services in 6G-IoT networks. Part A is the network architecture of ASPs employing edge servers to deploy AIGC models for providing AIGC services to clients; Part B shows an illustration of edge AIGC services and presents variation in the subjective utility of a client from the same text prompt on various ASPs.
• Figure 2: An illustration for the subjective utility of the client based on prospect theory in the contract theory model.
• Figure 3: Generative diffusion models for optimal contract design under prospect theory. Note that MLP refers to the multi-layer perception10409284.
• Figure 4: Test reward comparison of our scheme with other schemes under PT, i.e., contract-based incentive mechanism with complete information and random policy under asymmetric information, where reference point $U_{ref}=200$ and loss aversion $\eta = 0.5$.
• Figure 5: Performance comparison between the diffusion-based SAC algorithm and traditional DRL algorithms in optimal contract design under PT, where reference point $U_{ref}=200$ and loss aversion $\eta = 0.5$.

Theorems & Definitions (15)

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