floriank13 / abm-calibration

Agent based model of Innovation Diffusion with a focus on the calibration algorithm

Calibrating Agent-based Models of Innovation Diffusion with Gradients

The presented Agent based model (short: ABM) of innovation diffusion is a very basic ABM that can model the temporal and spatial parts of an Innovation diffusion process. The focus in this repository lies on the main feature, namely the gradients of the model. These gradients enable researchers in the field of Innovation Diffusion to calibrate their models more efficiently.

The model is part of the publication "Calibrating Agent-based Models of Innovation Diffusion with Gradients" published here. If you use the code from this repository in a scientific work, please cite the paper. A step-by-step explanation of the algorithm and an implementation can be found in the jupyter notebook code/ABM_adoption.ipynb.

Additionally there is a short summary of the paper on youtube.

If you find a bug or have suggestions for improvement, just open a new Issue.