Agentic AI Reference Architecture

Cognition · Software component

Monte Carlo Planner

Software componentCognitionCognition & Memoryarc:MonteCarloPlanner

A simulation-based task planner that samples and evaluates complete trajectories through a forward model without maintaining a search tree, discarding intermediate states after each iteration.

Responsibility. Plans by evaluating independently sampled full trajectories.

Also known as: Monte Carlo Planning (MCP), Tree-less Monte Carlo planning

Variant of Task Planner abstract

When to choose. Choose for one-shot planning problems with expensive forward-model simulations (e.g., computationally intensive physics), where lower memory use outweighs MCTS's progressive statistical refinement.

specializesis target of alternativeTois configured byinvokesalternative tois constrained byTask Planner: specializesTask PlannerMCTS Planner: is target of alternativeToMCTS PlannerReward Function Specification: is configured byReward Function Specific…Environment Simulator: invokesEnvironment SimulatorOptimal Heuristic Search Planner: alternative toOptimal Heuristic Search…Search Budget Policy: is constrained bySearch Budget Policy
Direct neighbourhood (hover for relationship types)

Relationships

is configured by structural

invokes dependency

is constrained by control

alternative to variability

Classification

Patterns
Monte Carlo planning without tree
Quality attributes
Performance efficiency (ISO/IEC 25010)Maintainability (ISO/IEC 25010)

Sources

  1. Ch5.5: T. Nguyen, "Monte Carlo Tree Search Fundamentals," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 5.5. ISBN: 9798244538229.