A Bayesian game decision-making model for uncertain adversary types
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
Adaptive application security involves making decisions under uncertainties such as the time, the power, or the damage of potential attacks. One of the uncertainties that has been largely ignored in the literature is the intention of the adversaries. The majority of research focuses on characteristics of attacks (e.g., their request arrival rates), whereas characteristics of attackers/adversaries (e.g., their intentions and strategies) are neglected. In today's sophisticated attacks, in order to confuse defense systems, adversaries may initiate an attack that exhibits a scenario similar to another attack but has an entirely different malicious goal (e.g., to break down the server or to harm a specific user in the system). In such cases, incorporating uncertainty about the type of adversaries into the decision model helps to choose a proper countermeasure for protecting the software system efficiently. In this paper, we present a Bayesian game model that captures the uncertainty about an adversary's motivation for sending malicious requests. Our game-theoretic model formalizes possible intentions of adversaries along with the security preferences of the software system. In such a novel design, the equilibrium of the modeled game balances the gain from achieving security goals with the loss incurred by mitigating the attack. We provide an extensive analysis of the proposed game-theoretic model in the presence and absence of uncertainty about the adversary type. Moreover, we present a case study to show how such uncertainty can be addressed using the proposed technique in a real-world scenario.
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 | 0.001 |
| 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 | 0.001 |
| Open science | 0.001 | 0.000 |
| 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