Emergent communication in MARL — canonical result.
hardAnswer
- Agents can learn to communicate over discrete tokens via reward gradient (differentiable via Gumbel-softmax or reinforce).
- Foerster et al.'s DIAL demonstrated on switch-riddle: agents develop protocol to signal correct answer.
- Later work: emergent grammar and compositionality (with pressure).
- Modern: LLM-agent negotiation and collaboration extends emergent-comm to natural language.
Check yourself — multiple choice
- Random
- Agents learn communication over discrete tokens (Gumbel-softmax / reinforce); Foerster's DIAL on switch-riddle; extends to emergent grammar / natural-lang LLM agents
- Same as PPO
- Not real
Emergent comm: learn protocols via reward gradients; extends to LLM-agent NL.
#multi-agent
Practise Reinforcement Learning
214 interview questions in this topic.