ProtoDINet: End-to-End Interpretable Prototypical Model for Insulator Damage Detection
Bibliographic record
Abstract
Damaged grid components such as insulators in transmission line can cause widespread impact in downstream connections such as outages. Preventing such failures requires frequent laborious inspection of thousands of power grid components, hence currently is done quite infrequently. In an era of societal electrification, there is an increasing need for an automated approach to continuous monitoring and inspection of grid components. In this paper, we propose an explainable deep learning based approach for damage detection and automated inspection of transmission lines from aerial images. We propose ProtoDINet - an interpretable end-to-end trained prototypical damaged insulator detection network. Unlike the black box machine learning models, ProtoDINet learns explainable prototypes during training-making it inherently interpretable. The most relevant prototypes are then automatically identified using a combinatorial multi-armed bandit (MAB) method targeted to improve class distinction. The final learned prototypes facilitate end users’ understanding of the model’s decision making process. We report promising results in F1 and AUC scores, visualize meaningful learnt prototypes, and demonstrate the effectiveness of the MAB-based prototype selection both quantitatively and qualitatively. This demonstrates the potential for more trustworthy interpretable solutions compared to existing black-box methods and encourages further research in the direction of prototypical networks for power grid.
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.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.004 | 0.001 |
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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".