Selective Reincarnation introduces a dynamic population-based training mechanism tailored for multi-agent reinforcement learning. Each agent’s performance is regularly evaluated against predefined thresholds. When an agent’s performance falls below its peers, its weights are reset to those of the current top performer, effectively reincarnating it with proven behaviors. This approach maintains diversity by only resetting underperformers, minimizing destructive resets while guiding exploration toward high-reward policies. By enabling targeted heredity of neural network parameters, the pipeline reduces variance and accelerates convergence across cooperative or competitive multi-agent environments. Compatible with any policy gradient-based MARL algorithm, the implementation integrates seamlessly into PyTorch-based workflows and includes configurable hyperparameters for evaluation frequency, selection criteria, and reset strategy tuning.

Patrolling-Zoo offers a flexible framework enabling users to create and experiment with multi-agent patrolling tasks in Python. The library includes a variety of grid-based and graph-based environments, each simulating surveillance, monitoring, and coverage scenarios. Users can configure the number of agents, map size, topology, reward functions, and observation spaces. Through compatibility with PettingZoo and Gym APIs, it supports seamless integration with popular reinforcement learning algorithms. This environment facilitates benchmarking and comparing MARL techniques under consistent settings. By providing standard scenarios and tools to customize new ones, Patrolling-Zoo accelerates research in autonomous robotics, security surveillance, search-and-rescue operations, and efficient area coverage using multi-agent coordination strategies.