Fractal based cognitive neural network to detect obfuscated and indistinguishable internet threats
Bibliographic record
Abstract
State of the art network intrusion detection systems are heavily influenced by signature based techniques for detecting threats which are extracted from raw packet captures and firewall logs. With the recent emergence of cloud computing and big data analytics, supervised machine learning is also being used to detect deviations of the network traffic patterns from already-known normal patterns. Subsequently, these anomalies are analyzed by human experts to differentiate legitimate anomalies, also known as true positives, from enormous false anomalies and reconfigure the machine learning system accordingly. Using machine learning for cyber security is relatively a difficult topic compared to other application domains primarily because of the dynamically fast changing threat landscape which is also extremely complex. Our main claim is that the proposed methodology significantly improves the classification performance of neural networks by detecting obfuscated malicious samples that masquerade the behavior of normal samples and thus are indistinguishable on Platonic Euclidean feature space. It is achieved by transforming a traditional single-scale (Euclidean scale) based error curve to information fractal dimension based multiscale error curve and subsequent design changes in the backpropagation algorithm. The performance comparison is provided by incorporating our proposed methodology in a fundamental gradient descent based neural network and shows promising results. Our claims are supported by experimental results and the subsequent analyses.
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 | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| 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".