Robust Federated Learning for Mitigating Advanced Persistent Threats in Cyber-Physical Systems
Bibliographic record
Abstract
Malware triage is essential for the security of cyber-physical systems, particularly against Advanced Persistent Threats (APTs). Proper data for this task, however, are hard to come by, as organizations are often reluctant to share their network data due to security concerns. To tackle this issue, this paper presents a secure and distributed framework for the collaborative training of a global model for APT triage without compromising privacy. Using this framework, organizations can share knowledge of APTs without disclosing private data. Moreover, the proposed design employs robust aggregation protocols to safeguard the global model against potential adversaries. The proposed framework is evaluated using real-world data with 15 different APT mechanisms. To make the simulations more challenging, we assume that edge nodes have partial knowledge of APTs. The obtained results demonstrate that participants in the proposed framework can privately share their knowledge, resulting in a robust global model that accurately detects APTs with significant improvement across different model architectures. Under optimal conditions, the designed framework detects almost all APT scenarios with an accuracy of over 90 percent.
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 | 0.003 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 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".