MARL to Choose Actions On-the-Fly in a Cognitive Radar System
Bibliographic record
Abstract
Task scheduling for radar systems have generally taken a block approach, scheduling a given set of tasks, with predetermined task parameters, to optimize some cost function. This paper takes a contrasting, on-the-fly, approach, determining what the radar(s) must “do next” to maximize their understanding of their operating environment. Specifically, we develop reinforcement learning (RL) for joint tracking and surveillance in a multi-function radar system. We further develop a multi-agent reinforcement learning (MARL) system to coordinate the actions of two radars. In this paper, we define actions as the act of viewing an azimuth around a radar. We outline the system's architecture, environmental dynamics, RL agents employed, and comprehensive metrics for evaluating agents' performance. Our approach presents significant challenges due to the ambiguity in defining a state space, the extensive time required to train agents, and the large number of tunable RL hyperparameters. Additionally, the fundamentally opposed nature of tracking and searching tasks introduces a dichotomy in the requirements of a reward function for the RL agent. The findings of this paper provide insight and advancements in the field of cognitive radar systems.
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 | 0.001 | 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".