Agent Dynamics
Python experiments for interacting agent models.
Completion: Experiment / prototype · Lean: No Lean claim · Confidence: Experimental evidence
Before you start
Suggested background: Basic programming and familiarity with the stated data and evaluation method. Read the input limits and acceptance criteria before interpreting output.
Recorded results and limits: Executed illustrative two-agent model with supplied schedule, sensor semantics and reset rule. No learned policy or general self-improvement result.
Remaining work: Validate any learned or adaptive replacement against a preregistered baseline. Retain the supplied-rule demonstration as its own experiment.
The background advice is an editorial starting point. The work itself states its prerequisites and hypotheses.
Reading routes · Glossary · Claim links · Download this work
Read
Written proofs
Read the written arguments in the PDF. The manuscript source remains available below.
Find the statement number or section in the reading files, then read its assumptions and argument. A claim map may cover only part of the written result.
Written proofs may be in the reading files. There is no separate text or claim map linked here. The evidence record describes what has been reviewed.
Find this work among other written proofs
Lean
No Lean source or guide is linked here. See the Lean status above and the evidence below for the recorded coverage.
Code
These links include code, examples, checks, or build tools. Read the edition guide for setup instructions and what each command does. All software is beta and work in progress.
Find this work among other code projects
Evidence
- Completion, Lean coverage, confidence, and cited evidence
- Source-file receipt (identity and availability)
- All files in this source edition
The evidence records describe the reviews and runs completed for this edition.