Deep Neural Network Guided Evolution of L-System Trees
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
Lindenmayer systems (L-systems) are mathematical formalisms used for generating recursive structures. They are particularly effective for defining realistic tree and plant models. It takes experience to use L-systems effectively, however, as the final rendered results are often difficult to predict. This research explores the use of genetic programming (GP) and deep learning towards the automatic evolution of L-system expressions that render 2D tree designs. As done before by other researchers, the L-system language is easily defined and manipulated by the GP system. It is challenging, however, to determine a fitness function to evaluate the suitability of evolved expressions. We train a deep convolutional neural network (CNN) to recognize suitable trees rendered in the style of the L-system language. Experiments explore a number of deep CNN strategies. Results in some experiments are very promising, as images conforming to specified styles of tree species were often produced. We found that underspecifying or over-complicating the training requirements can arise, and the results become unsatisfactory in such cases. Our results also confirm that of other researchers, in that deep learning can be fooled by evolutionary algorithms, and the criteria for success learned by deep neural networks might not conform with those of human users.
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 | 0.000 |
| 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