Collusion-Resistant Privacy-Preserving Outsourced Training Under Single Cloud With Semi-Honest TEE
Bibliographic record
Abstract
In the era of big data, leveraging multi-party data for model training has become essential to overcome the limitations of single-source data. However, the integration and processing of data are restricted by stringent laws and regulations to ensure privacy. Although some privacy-preserving multi-party training schemes have been proposed, they still face limitations in computation, communication, or security. In this paper, we propose PACLR and PACLR$^+$to achieve collusion-resistant privacy-preserving outsourced training under single cloud with semi-honest TEE. PACLR protects data security, while PACLR$^+$ensures the security of both data and model parameters. Specifically, we introduce a trust-enhanced and secure outsourcing training approach that eliminates the need for a trusted third party to initialize the system and achieves secure data outsourcing. Then, we present a privacy-preserving training method over packed ciphertext using mini-batch stochastic gradient descent to balance efficiency and accuracy. Additionally, we propose some optimization approaches to improve training efficiency. Detailed security analysis confirms that our schemes are privacy-preserving and can resist various collusion attacks. Extensive experiments on real datasets demonstrate the effectiveness of our schemes, showing superior efficiency compared to recent scheme.
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 | 0.004 | 0.016 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.002 | 0.006 |
| Open science | 0.003 | 0.008 |
| Research integrity | 0.002 | 0.003 |
| Insufficient payload (model declined to judge) | 0.002 | 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".