Paweł Romańczuk

This study tracks human foragers and their prey to show adaptive predator-prey dynamics, including foraging strategies and prey responses to human activity.

This paper develops a computational framework for distributional multi-agent reinforcement learning, showing how optimism, reward discretization, and temporal discounting interact to shape decision-making.

This study shows that the quality of social information and environmental volatility influence collective foraging strategies in a multi-agent model.

The paper shows larger fish shoals in the wild better detect predator attacks while maintaining constant false alarms and faster decision times.

This study introduces CoBe XR, a system enabling real-time human interaction with virtual swarms through natural movement, showing participants can adapt to and control swarm dynamics without specialized interfaces.

The paper investigates collective escape patterns in sardines under predator attack, showing fountain effects arise from optimal escape rules and predator strategies.

This study shows that moderate confirmation bias improves collective decision-making in reinforcement-learning agents, but excessive bias can lead to suboptimal performance and polarization.

This study presents an agent-based model to explore how individual decision-making based on personal and social information influences collective foraging outcomes in different environments.