The spatially-explicit AgriculTuralLandscApe Simulator (ATLAS) simulates realistic spatial-temporal crop availability at the landscape scale through crop rotations and crop phenology.

This model explores different aspects of the formation of urban neighbourhoods where residents believe in values distant from those dominant in society. Or, at least, this is what the Danish government beliefs when they discuss their politics about parallel societies. This simulation is set to understand (a) whether these alternative values areas form and what determines their formation, (b) if they are linked to low or no income residents, and (c) what happens if they disappear from the map. All these three points are part of the Danish government policy. This agent-based model is set to understand the boundaries and effects of this policy.

This model is a replication of Torsten Hägerstrand’s 1965 model–one of the earliest known calibrated and validated simulations with implicit “agent based” methodology.

C++ and Netlogo models presented in G. Bravo (2011), “Agents’ beliefs and the evolution of institutions for common-pool resource management”. Rationality and Society 23(1).

This repository the multi-agent simulation software for the paper “Comparison of Competing Market Mechanisms with Reinforcement Learning in a CarPooling Scenario”. It’s a mutlithreaded Javaapplication.

The BENCH agent-based model is designed and developed to study shifts in residential energy use and corresponding emissions driven by behavioral changes among heterogeneous individuals.

3 simple models to illustrate diffusion of innovations.
The models are discussed in Introduction to Agent-Based Modeling by Marco Janssen. For more information see https://intro2abm.com/

This is based off my previous Profiler tutorial model, but with an added tutorial on converting it into a model usable with BehaviorSpace, and creating a BehaviorSpace experiment.

A discrete-time stochastic model with state-dependent transmission probabilities and multi-agent simulations focusing on possible risks that could materialize in the final phase of the epidemic.

This model implements a classic scenario used in Reinforcement Learning problem, the “Cliff Walking Problem”. Consider the gridworld shown below (SUTTON; BARTO, 2018). This is a standard undiscounted, episodic task, with start and goal states, and the usual actions causing movement up, down, right, and left. Reward is -1 on all transitions except those into the region marked “The Cliff.” Stepping into this region incurs a reward of -100 and sends the agent instantly back to the start (SUTTON; BARTO, 2018).

The problem is solved in this model using the Q-Learning algorithm. The algorithm is implemented with the support of the NetLogo Q-Learning Extension