Learnability of Competitive Threshold Modles

• Develop theoretical tools for analyzing sample complexity and generalization bounds for learning competitive diffusion models.
• Design a realizable hypothesis space for learning competitive diffusion models. We explicitly simulate the competitive diffusion threshold models by artificial neural networks with finite V.C. dimensions.
• Present theoretical analysis to analyze the PAC learnability of the present learning problem, as well as sample complexity and generalization bounds for learning competitive diffusion models.
• Formulate the efficient learning algorithm via linear programming with the polynomial number of constraints.

Query-Decision Regression for Social Contagion Management

• Create a query-decision regression framework for social contagion management using structured prediction techniques and deep graph kernels.
• Develope a deep structure learning framework with gradient descent techniques for graph kernel functions to infer high-quality decisions for future queries.