AP2FL: Auditable Privacy-Preserving Federated Learning Framework for Electronics in Healthcare
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
The growing application of machine learning (ML) techniques in healthcare has led to increased interest in federated learning (FL), which enables the secure and private training of robust ML models. However, conventional FL methods often fall short of providing adequate privacy protection and face challenges in handling non-independent and identically distributed (Non-IID) training data. These shortcomings are of significant concern when employing FL in electronic devices in healthcare. To address these issues, we propose an Auditable Privacy-Preserving Federated Learning (AP2FL) model tailored for electronics in healthcare settings. By leveraging Trusted Execution Environments (TEEs), AP2FL ensures secure training and aggregation processes on both client and server sides, effectively mitigating data leakage risks. To manage Non-IID data within the proposed framework, we incorporate the Active Personalized Federated Learning (ActPerFL) model and Batch Normalization (BN) techniques to consolidate user updates and identify data similarities. Additionally, we introduce an auditing mechanism in AP2FL that reveals the contribution of each client to the FL process, facilitating the updating of the global model following diverse data types and distributions. In other words, it ensures the FL process’s integrity, transparency, fairness, and robustness. Our results demonstrate that the proposed AP2FL model outperforms existing methods in accuracy and effectively eliminates privacy leakage.
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 | 0.001 | 0.003 |
| Meta-epidemiology (narrow) | 0.000 | 0.001 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.004 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.011 | 0.001 |
| Research integrity | 0.001 | 0.003 |
| 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