Report 6: Diversity in Evolutionary Ensembles of Artificial Neural Networks 1 Summary
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.
Bibliographic record
Abstract
Since the last report, and on the advice received in the last meeting, the focus of this work has now narrowed to examining methods for creating diversity in evolutionary ensembles of neural networks. The three areas that will be examined in the thesis are diversity in local search and global evolution in the EENCL framework, and the investigation of a novel combination of NCL and a multi-objective evolutionary approach. This report details some of the results achieved and publications written since the last report and also details other work that is ongoing. A detailed thesis plan is presented, as well as a timetable to completion. It is expected that the experimental results will be completed within the funded period, but that writing up will overun the funded period by up to three months. 2 Results and developments Our recent results with experiments using an Island model population structure (INCL) are presented in two conference papers in appendix A and B. The first is a paper presented at CEC06 in Vancouver, detailing the INCL approach. The second is a paper submitted to ICDM06 and recently rejected, that attempted to expand on the earlier paper with some comparative and analytical work that was unavailable for the CEC paper. The paper was rejected mainly because the comparitive
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 imitationNot 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.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it