In October 2018, I talked about how allocation policies and participation behavior influence social welfare in platform economy at JADS. You can find my slides here:
Super excited and proud! Our paper “Learning optimal classification trees using a binary linear program formulation” was accepted by AAAI (Thirty-Third AAAI Conference on Artificial Intelligence)! Acceptance rate 16.2% over 7700 submissions!
We show the power of optimization methods for deriving interpretable machine learning models.
A paper “Learning decision trees with flexible constraints and objectives using integer optimization” has been accepted by CPAIOR 2017!
Abstract: Task allocation problems have focused on achieving one-shot optimality. In practice, many task allocation problems are of repeated nature, where the allocation outcome of previous rounds may influence the participation of agents in subsequent rounds, and consequently, the quality of the allocations in the long term. We investigate how allocation influences agents’ decision to participate using prospect theory, and simulate how agents’ participation affects the system’s long term social welfare. We compare two task allocation algorithms in this study, one only considering optimality in terms of costs and the other considering optimality in terms of primarily fairness and secondarily costs. The simulation results demonstrate that fairness incentivizes agents to keep participating and consequently leads to a higher social welfare.
Qing Chuan Ye and Yingqian Zhang. Participation behavior and social welfare in repeated task allocations. IEEE International Conference on Agents. 2016. to appear.
(Download paper here)
Keywords: Agent based modelling and simulation; Repeated task allocation; Prospect theory; Fairness;Participation behavior.
The paper Fair Task Allocation in Transportation is now online at Omega:
The author’s copy is available to download here: Download paper
Introduction: In this paper we study a fair task allocation problem in transportation where an optimal allocation not only has low cost but more importantly, it distributes tasks as even as possible among heterogeneous participants who have different capacities and costs to execute tasks. We develop an efficient method to find the optimal solution to this problem. We show by experiments that fairness often comes with a very small price in terms of cost.
The motivation of this study comes from the fact that when dealing with task or resource allocation, more and more attention is being given to cases in which cost should not always be the sole or major consideration. This is true for many problems especially those in sharing economy.
We have a PhD position in the area of data-aided decision support, which combines machine learning/data mining techniques with optimisation for dynamic online markets. The deadline of the application is April 17. Please visit the following page for more details: