Skip to content

Simulation Lab

Agent Intelligence

Multi-agent worlds probing when communication, cooperation, and strategy arise.
← All categories

Emergent Agents

Multi-agent reinforcement-learning sandbox studying when communication and cooperation emerge from sparse rewards.

Running

Parameters

{
  "agents": 64,
  "reward": "sparse",
  "environment": "foraging"
}

Recorded runs (1)

run 51b00003started 15 Mar 2026in progress
{
  "cooperation_index": 0.71,
  "communication_emerged": true
}