Experimental Verification of a Convolutional Neural Network Separation Method in TeV Gamma-Ray Observations by the Tibet ASγ Experiment
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
Abstract Since 1990, the Tibet AS$\gamma$ experiment has been observing gamma rays and cosmic rays with energies greater than several TeV using a surface air shower array. An underground muon detector (MD) array operating since 2014 enables us to significantly discriminate between gamma rays and cosmic rays by counting the number of muons in the air showers. However, discrimination with only the air shower array is challenging. We developed a convolutional neural network (CNN)-based method to improve the sensitivity of gamma-ray measurement data recorded by only the air shower array. The area-under-the-curve values of the CNN method for gamma rays generated by a Monte Carlo (MC) simulation assuming a gamma-ray source (Crab Nebula) were 0.75 at $\sim$10 TeV and 0.83 at $\sim$100 TeV. The detection significances of gamma rays were improved by factors of 1.232 $\pm$ 0.007 at $\sim$10 TeV and 1.557 $\pm$ 0.022 at $\sim$100 TeV. For verification, we applied the proposed method to experimental data including high-purity gamma-ray-like events in the direction of the Crab Nebula, acquired using both arrays. The distributions of gamma-ray-like properties obtained from the CNN were in good agreement with the MC Simulation, with reduced $\chi ^2$ values of 0.507–1.57, corresponding to an upper cumulative probability of 0.120–0.871.
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 | 0.001 |
| 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