EasyDeepLearn

MADDPG — Multi-Agent DDPG.

hard

Answer

  • Lowe et al. 2017.
  • Each agent has its own actor πi(oi){\pi}_{i}(o_{i}) and critic Qi(s,  a1,  ,  an)Q_{i}(s, \;a_{1}, \;, \;a_{n}) (centralized: sees ALL agents' actions).
  • Actor uses only local obs, critic uses joint info → CTDE.
  • Handles cooperative + competitive + mixed.
  • Foundational cooperative-continuous-action MARL algorithm.
  • Modern variants: MATD3, MASAC.
Check yourself — multiple choice
  • Same as DDPG
  • Each agent: local actor πi(oi)  +  centralized{\pi}_{i}(o_{i})\; + \;\mathrm{centralized} critic Qi(s,  a1,  ,  an)Q_{i}(s, \;a_{1}, \;, \;a_{n}); CTDE for cooperative + competitive + mixed; MATD3 / MASAC modern variants
  • Random
  • Not real

MADDPG: local actor + centralized critic; CTDE cooperative-mixed.

#multi-agent#actor-critic

Practise Reinforcement Learning

214 interview questions in this topic.

Read the lessonFlashcardsTake the quiz

Related questions