Meta-ATMoS+: A Meta-Reinforcement Learning Framework for Threat Mitigation in Software-Defined Networks
Bibliographic record
Abstract
As cyber threats become increasingly common, automated threat mitigation solutions are more necessary than ever. Conventional threat mitigation frameworks are difficult to tune for different network environments, but frameworks utilizing deep reinforcement learning (RL) have been proven to be an effective approach that can adapt to different networks automatically. Existing RL-based frameworks have shown to be generalizable to different network sizes and threats, and robust to false positives. However, training RL agents for these frameworks can be challenging in a production environment as the training process is time-consuming and disruptive to the production network. Hence, a staging environment is required to effectively train them. In this paper, we propose Meta-ATMoS+, a meta-RL framework for threat mitigation in software-defined networks. We leverage Model-Agnostic Meta-Learning (MAML) to find an initialization for the RL agent that generalizes to a variety of different network configurations. We show that the RL agent with MAML-learned initialization can accomplish few-shot learning on a target network with comparable performance to training on a staging environment. Few-shot learning not only allows the model to be trainable directly in the production environment but also enables human-in-the-loop RL for the mitigation of threats that do not have an easily-definable reward function.
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 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| 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 | 0.002 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.003 | 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".