Using machine learning to interpret 3D airborne electromagnetic inversions
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
SummaryAlthough 3D airborne electromagnetic inversions have improved greatly in recent years, the presence of smooth boundaries has often been a strong criticism. This smoothness can easily be remedied by applying different types of regularization and constraints to the model, but another approach is to learn what underlying structures or boundaries these smooth transitions represent.To perform this advanced inversion interpretation, we trained a machine learning algorithm known as VNet to identify the relationship between a true synthetic model and the resulting smooth 3D inversion model. By training on one section of the model and predicting on another, the algorithm was able to learn the general relationships required to intelligently sharpen the inversion model in the prediction area. The resulting images approximate the true synthetic model to a much closer degree compared to the original inversion model. The VNet was trained in two ways, one to predict a conductivity value for each pixel, and another to predict a classification unit for each pixel presuming the conductivity for each class is known. Each method performed similarly well with some minor differences, which gives the user some options depending on the scenario and how much a priori information is known.Overall this automatic interpretation technique worked well over a synthetic model, and future simulations will be run in order to make the method more robust and applicable for field scenarios.
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 | 0.000 |
| 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) | 0.004 | 0.003 |
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