MétaCan
Menu
Back to cohort

Think Before You Act: Popperian Expectations for Adaptive Agents

2025· article· W4416184268 on OpenAlex

Why this work is in the frame

A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.

affAt least one author lists a Canadian institution in the pinned OpenAlex snapshot.

Bibliographic record

Venuenot available
Typearticle
Language
FieldComputer Science
TopicAI-based Problem Solving and Planning
Canadian institutionsOntario Tech University
Fundersnot available
KeywordsMeaning (existential)Reinforcement learningInternal modelMechanism (biology)Event (particle physics)Autonomous agentCausal modelValue (mathematics)

Abstract

fetched live from OpenAlex

Autonomous agents that operate and learn in complex environments must continually balance multiple objectives, including efficiency and risk. Trial-and-error reinforcement learning can reduce the efficiency and safety of agents due to the need to incorporate exploration along with the exploitation of learnt knowledge. Internal simulation frameworks, such as Winfield's consequence engine, have demonstrated the value of using internal simulations in agents to predict the outcomes of actions without requiring physical commitment. However, the framework lacks a formal mechanism for storing knowledge learned from these internal simulations, meaning agents are often stuck repeatedly simulating scenarios they have encountered previously. This paper introduces Popperian Expectations and a novel architecture that extends Winfield's framework by enabling agents to create, update, and utilise causal expectations learned from internal simulations. Using the Expectation Event Calculus (EEC), the agent can form interpretable causal expectations based on internal simulations. This paper explores how Popperian Expectations can be used to enable agents to reason reflectively and adapt to complex environments.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.000
metaresearch head score (Gemma)0.000
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesMeta-epidemiology (narrow)
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: none
GenreCandidate signal: Methods · Consensus signal: none
Teacher disagreement score0.934
Threshold uncertainty score1.000

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0000.000
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.001
Science and technology studies0.0010.000
Scholarly communication0.0010.001
Open science0.0010.000
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0000.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.030
GPT teacher head0.292
Teacher spread0.262 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Quick stats

Citations2
Published2025
Admission routes1
Has abstractyes

Explore more

Same topicAI-based Problem Solving and PlanningFrench-language works237,207