teleo/teleo-codex
8
0
Fork 0
Code Issues 22 Pull requests 201 Projects Releases Packages Wiki Activity Actions 1

theseus: extract claims from 2015-03-00-friston-active-inference-epistemic-value #181

Merged
leo merged 1 commit from extract/2015-03-00-friston-active-inference-epistemic-value into main 2026-03-10 19:37:37 +00:00
Conversation 2 Commits 1 Files changed 1 +6 -1
1 changed files with 6 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files
Showing only changes of commit bcf1935a07 - Show all commits

7
inbox/archive/2015-03-00-friston-active-inference-epistemic-value.md
View file

@ -7,9 +7,14 @@ date: 2015-03-00
domain: ai-alignment domain: ai-alignment
secondary_domains: [collective-intelligence, critical-systems] secondary_domains: [collective-intelligence, critical-systems]
format: paper format: paper
status: unprocessed status: null-result
priority: high priority: high
tags: [active-inference, epistemic-value, information-gain, exploration-exploitation, expected-free-energy, curiosity, epistemic-foraging] tags: [active-inference, epistemic-value, information-gain, exploration-exploitation, expected-free-energy, curiosity, epistemic-foraging]
processed_by: theseus
processed_date: 2025-03-10
enrichments_applied: ["structured-exploration-protocols-reduce-human-intervention-by-6x-because-the-Residue-prompt-enabled-5-unguided-AI-explorations-to-solve-what-required-31-human-coached-explorations.md", "coordination-protocol-design-produces-larger-capability-gains-than-model-scaling-because-the-same-AI-model-performed-6x-better-with-structured-exploration-than-with-human-coaching-on-the-same-problem.md"]
extraction_model: "anthropic/claude-sonnet-4.5"
extraction_notes: "Foundational paper on epistemic value in active inference. Extracted three claims: (1) epistemic foraging as Bayes-optimal behavior, (2) deliberate vs habitual mode governed by uncertainty, (3) confirmation bias as signal of suboptimal foraging. Enriched two existing claims about structured exploration protocols with theoretical grounding from active inference framework. All three new claims are immediately operationalizable for agent architecture: epistemic value targeting, domain maturity assessment, confirmation bias detection."
--- ---
## Content ## Content