Stackelberg Security Game with Reinforcement Learning for Defending Video Servers Against APT Attacks
Bibliographic record
Abstract
Advanced Persistent Threats (APTs) pose a significant challenge for video servers in critical infrastructures. APT attacks tend to be multi-stage, difficult to attack for traditional intrusion detection systems (IDS) with advanced evasion techniques to evade modern machine learning-based defences. To address these challenges, we introduce SSG-RL. This hybrid defence framework integrates Stackelberg Security Game (SSG), Advantage Actor-Critic Reinforcement Learning (A2C) and a Risk-Rank (RR) algorithm, implemented within a Network Digital Twin (NDT) using the Cyber Security Learning Environment (CSLE). SSG-RL jointly models attacker tactics and defender strategies, enabling proactive responses mapped to the common MITRE ATT&CK and MITRE D3FEND frameworks. Validated using emulation traces from an external testbed, we demonstrate that SSG-RL enhances detection accuracy and reduces response latency compared to three baseline techniques: one that utilises a traditional IDS and two reinforcement learning approaches (DQN and PPO). These results highlight the potential of combining game-theoretic foresight with adaptive learning for scalable cyber defence.
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 | 0.005 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 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".