Summarizes the paper’s concrete rule-learning instantiation and clearly separates DE11 from transformers, language models, and browser runtimes.
Research documentation: this page interprets a cited research source and defines evidence requirements. It does not claim a released Teleodynamic AI implementation.
What it is
DE11 is the source paper’s concrete rule-learning instantiation. It uses interpretable logical Forms as hypotheses, continuous parameters, discrete structural actions, an energy variable, natural-gradient updates, and tropical selection.
Author-reported evidence
The paper reports 93.3% IRIS, 92.6% WINE, and 94.7% Breast Cancer accuracy, along with phase and halt analyses. MiRust has not independently reproduced the software, data splits, proofs, or results.
Important negative result
The paper reports 55.9% on DIGITS and treats this as evidence of scalability limits for the current rule representation.
Not equivalent to
DE11 is not TinyLM-16M, a transformer, a Small Language Model, a browser inference engine, or the GGUF implementation.
Scope
This starter page defines the questions, boundaries, evidence, and failure modes that should be recorded before a capability is presented as supported.
Engineering considerations
- Identify the source, version, target environment, and owner.
- Separate observed values from estimates and externally reported values.
- Record trade-offs, unsupported cases, and fallback behavior.
- Link performance statements to a compatible benchmark methodology.
Verification questions
- What exact artifact, revision, backend, and environment were reviewed?
- Which assumptions could change the result?
- Which data should be retained so another engineer can reproduce the conclusion?