Some challenges of calibrating differentiable agent-based models
This addresses the problem of applying ABMs in practice for researchers and practitioners, but it is incremental as it focuses on existing challenges without major breakthroughs.
The paper tackles the challenges of calibrating differentiable agent-based models (ABMs), which are complex and discrete, by discussing and experimenting with potential solutions to improve parameter inference and optimization.
Agent-based models (ABMs) are a promising approach to modelling and reasoning about complex systems, yet their application in practice is impeded by their complexity, discrete nature, and the difficulty of performing parameter inference and optimisation tasks. This in turn has sparked interest in the construction of differentiable ABMs as a strategy for combatting these difficulties, yet a number of challenges remain. In this paper, we discuss and present experiments that highlight some of these challenges, along with potential solutions.