Principles of metareasoning

1991/05/01 by Stuart Russell, Eric Wefald · 291 citations
Computer Science · Mathematics · #AI-based Problem Solving and Planning #Algorithm #Artificial intelligence #Basis (linear algebra) #Bounded rationality #Computation #Computational complexity theory #Computational resource #Computer science #Decision theory #Epistemology #Logic, Reasoning, and Knowledge #Management science #Mathematical economics #Mathematics #Multi-Agent Systems and Negotiation #Outcome (game theory) #Programming language #Rationality #Semantics (computer science) #Sketch #Theoretical computer science

paper · doi:10.1016/0004-3702(91)90015-c

published in Artificial Intelligence 49(1-3), 361-395 (Elsevier BV)

openalex publication_date 1991/05/01 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/25

Cited by

• Re-FORC: Adaptive Reward Prediction for Efficient Chain-of-Thought Reasoning
• When to Plan: Learning to Select Between Reactive Control and Deliberative Planning
• How Intrinsic Motivation Underlies Embodied Open-Ended Behavior
• Automatically Composing Representation Transformations as a Means for Generalization
• Metacognition Should Be the Scientific Framework for Bounded and Effective Self-Governance in Generative AI
• Selecting Computations: Theory and Applications
• Imagining and building wise machines: the centrality of AI metacognition
• Metareasoning for Planning Under Uncertainty
• Who Sees What? Structured Thought-Action Sequences for Epistemic Reasoning in LLMs
• A General Framework for Constrained Bayesian Optimization using Information-based Search
• A Comprehensive Overview and Survey of Recent Advances in Meta-Learning
• Towards Rational Deployment of Multiple Heuristics in A*
• Truly Self-Improving Agents Require Intrinsic Metacognitive Learning
• BADGER: Learning to (Learn [Learning Algorithms] through Multi-Agent Communication)
• Meta-reasoning Using Attention Maps and Its Applications in Cloud Robotics
• RLKD: Distilling LLMs' Reasoning via Reinforcement Learning
• TemplateRL: Structured Template-Guided Reinforcement Learning for LLM Reasoning
• When Replanning Becomes the Bottleneck: Budgeted Replanning for Embodied Agents
• Humans Disengage, Reasoning Models Persist: Separating Difficulty Registration from Deliberation Allocation
• Scalable Dynamic Distributed Constraint Optimization with Metareasoning and Application to Continual Satellite Operations
• Rational metareasoning and the plasticity of cognitive control. [europepmc]
• Sensorimotor priors are effector dependent. [europepmc]
• Retrospective Inference as a Form of Bounded Rationality, and Its Beneficial Influence on Learning. [europepmc]
• Fixation patterns in simple choice reflect optimal information sampling. [europepmc]
• Deep imagination is a close to optimal policy for planning in large decision trees under limited resources. [europepmc]
• A Validated Ontology for Metareasoning in Intelligent Systems. [europepmc]
• Decision, Inference, and Information: Formal Equivalences Under Active Inference. [europepmc]
• A Resource-Rational Account of Human Eye Movements During Immersive Visual Search. [europepmc]