Internet of Things Intrusion Detection: Centralized, On-Device, or Federated Learning?
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
With the ever increasing number of cyber-attacks, internet of Things (ioT) devices are being exposed to serious malware, attacks, and malicious activities alongside their development. While past research has been focused on centralized intrusion detection assuming the existence of a central entity to store and perform analysis on data from all participant devices, these approaches cannot scale well with the fast growth of ioT connected devices and introduce a single-point failure risk that may compromise data privacy. Moreover, with data being widely spread across large networks of connected devices, decentralized computations are very much in need. in this context, we propose in this article a Federated Learning based scheme for ioT intrusion detection that maintains data privacy by performing local training and inference of detection models. in this scheme, not only privacy can be assured, but also devices can benefit from their peers' knowledge by communicating only their updates with a remote server that aggregates the latter and shares an improved detection model with participating devices. We perform thorough experiments on an NSL-KDD dataset to evaluate the efficiency of the proposed approach. Experimental results and empirical analysis explore the robustness and advantages of the proposed Federated Learning detection model by reaching an accuracy close to that of the centralized approach and outperforming the distributed unaggregated on-device trained models.
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 | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| 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