Efficient Energy Theft Detection Utilizing Hierarchical Federated Learning
Bibliographic record
Abstract
Energy theft in smart grids is a widespread challenge that impacts utility providers and consumers. It undermines the financial stability of utility companies, introduces electrical safety risks, and contributes to increased energy costs. The advancement of machine learning (ML) combined with the amount of data produced by smart grids can help resolve the problem by building efficient detection ML models. Data privacy is a hugely controversial issue in this case. Federated learning (FL) can be used to rescue as it enables the training of ML models while preserving data privacy by keeping data localized. However, it suffers compatibility issues, high communication costs, security threats, and scalability challenges. To address these limitations, this paper proposes to use hierarchical federated learning (HFL) for energy theft detection, where intermediate aggregations occur at edge servers, resolving compatibility issues, significantly reducing communication costs and enhancing scalability. It demonstrates how the novel HFL approach can efficiently manage and process data from distributed sources. Our experimental results demonstrate an 88% reduction in communication overhead and high scalability compared to FL. In terms of accuracy and computational efficiency, our approach is comparable to both FL and centralized ML.
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 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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 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".