Published in IAAI2021, 1900
We used machine learning models to predict the threats and then use a novel mixed-integer linear programming to plan the route in a mix of driving and walking patrols.
Published in IJCAI-PRICAI 2020, 2020
We introduced a new player – a strategic informant, who can observe and report upcoming attacks – to the defender-attacker security game setting.
Download here
Published in AAAI2020, 2020
We proposed a novel bi-level actor-critic learning method for multi-agent reinforcement learning that allows agents to have different knowledge base, while their actions still can be executed simultaneously and distributedly, and result in Stackelberg equilibrium as the solution.
Download here