AESMOTE: Adversarial Reinforcement Learning with SMOTE for Anomaly Detection
Bibliographic record
Abstract
Intrusion Detection Systems (IDSs) play a vital role in securing today's Data-Centric Networks.In a dynamic environment that is vulnerable to various types of attacks, novel, fast, and robust solutions are in demand to handle fast changing threats and thus the ever-increasing difficulty of detection.In this dissertation, we present a novel reinforcement learning based anomaly detection algorithm that further enables anomaly-based intrusion detection.As anomaly detection frameworks are mostly supervised learning based, which seeks the advantage of stable predictions and good performance with pre-recorded datasets, we have further explored the performance of a combined framework of joining a reinforcement learning algorithm with classimbalance techniques.The motivation of this approach is to not only exploit the auto-learning ability from the reinforcement learning loop, but also correct the classimbalance problem, which is pervasive in existing solutions.Our proposed solution is developed based on AE-RL [1].We further introduce an adapted SMOTE to address the class-imbalance problem while remodel the behaviors of the environmental agent for better performance.Experiments are conducted using NSL-KDD [2] datasets.Comparative evaluation and their results are presented and analyzed.Using techniques such as SMOTE, ROS, NearMiss1 and NearMiss2, performance measures obtained from our simulations have led us to recognize specific performance trends.The proposed model AESMOTE outperforms the original AE-RL in several cases.Experiment results show an Accuracy greater than 0.82 and F1 greater than 0.824.
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 | 0.002 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| 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".