Multi-layer stacking ensemble learners for low footprint network intrusion detection
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
Abstract Machine learning has become the standard solution to problems in many areas, such as image recognition, natural language processing, and spam detection. In the area of network intrusion detection, machine learning techniques have also been successfully used to detect anomalies in network traffic. However, there is less tolerance in the network intrusion detection domain in terms of errors, especially false positives. In this paper, we define strict acceptance criteria, and show that only very few ensemble learning classifiers are able to meet them in detecting low footprint network intrusions. We compare bagging, boosting, and stacking techniques, and show how methods such as multi-layer stacking can outperform other ensemble techniques and non-ensemble models in detecting such intrusions. We show how different variations on a stacking ensemble model can play a significant role on the classification performance. Malicious examples in our dataset are from the network intrusions that exfiltrate data from a target machine. The benign examples are captured by network taps in geographically different locations on a big corporate network. Among hundreds of ensemble models based on seven different base learners, only three multi-layer stacking models meet the strict acceptance criteria, and achieve an F1 score of 0.99, and a false-positive rate of 0.001. Furthermore, we show that our ensemble models outperform different deep neural network models in classifying low footprint network intrusions.
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 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.002 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| 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