Suggest edit — FV-POMCPs

Title

Name

Note

---
title: "FV-POMCPs"
type: concept
related: [Baysian Network, Maximum Likelihood Parameter Learning, Pomcp]
source: https://www.jemoka.com/posts/kbhfv_pomcps/
confidence: high
status: active
---

Main problem: joint actions and observations are exponential by the number of agents.
Solution: Smaple-based online planning for multiagent systems. We do this with the factored-value POMCP.
factored statistics: reduces the number of joint actions (through action selection statistics) factored trees: reduces the number of histories Multiagent Definition \(I\) set of agents \(S\) set of states \(A_{i}\) set of states for each agent \(i\) \(T\) state transitions \(R\) reward function \(Z_{i}\) joint observations for each agents \(O\) set of observations Coordination Graphs you can use sum-product elimination to shorten the Baysian Network of the agent Coordination Graphs (which is how agents influnece each other).
Mixture of Experts Directly search for the best joint actions; computed by MLE of the total value.