A Novel Distributed Multi-Agent Reinforcement Learning Algorithm Against Jamming Attacks
Bibliographic record
Abstract
In a multi-user wireless network under jamming attacks, self-interested users trying to learn their best anti-jamming strategy encounter a major problem, which is interference. An interesting solution is for users to learn anti-jamming techniques cooperatively, which can be achieved using a distributed learning algorithm. In this context, works proposing distributed learning algorithms to overcome jamming in multi-user applications, rely on the availability of a safe communication link to exchange information between users. However, if this communication link wireless, assuming that this link is safe against jamming attacks would not be accurate. Consequently, we propose a novel distributed multi-agent reinforcement learning algorithm for anti-jamming, namely Cross-Check Q-learning, where users build estimates of each other’s decision-making policies and adapt to them, therefore eliminating their need to communicate. When applied against both sweeping and smart jammers, our algorithm provides users with a better understanding of their environment and helps them learn the attacker’s policy and effectively avoid mutual interference. The proposed method’s transmission rates and interference levels is compared with the standard Q-learning, the collaborative multi-agent algorithm, and random policy. Simulation results show that our algorithm improves the users’ transmission rates, eliminates mutual interference, and has the highest convergence speed under the two considered jamming attacks.
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 | 0.001 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| 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".