Integrating Multimodality and Partial Observability Solutions Into Decentralized Multiagent Reinforcement Learning Adaptive Traffic Signal Control
Bibliographic record
Abstract
Adaptive Traffic Signal Control (ATSC) systems leverage sensor data to dynamically adjust signal timings based on real-time traffic conditions but they often suffer from partial observability (PO) due to sensor limitations and restricted detection ranges. This study addresses PO in fully decentralized ATSC systems by introducing eMARLIN-T, a controller designed to enhance performance by incorporating historical information in the decision-making process. Additionally, ATSC systems are commonly optimized to improve the performance of the general traffic, ignoring the impact on transit. On the other hand, traditional transit signal priority (TSP) strategies, which overlay preferential strategies for transit vehicles onto general traffic fixed signal plans, often lead to negative impacts on the general traffic. Thus, this paper tackles the challenge of optimizing traffic signals to benefit both public transit and general vehicular traffic. To address this, a novel decentralized multimodal multiagent reinforcement learning (RL) signal controller, eMARLIN-T-MM, is developed. This controller integrates a transformer-based encoder for transforming the state observations into a latent space and an executor Q-network for decision-making. Tested on a simulation of five intersections in North York, Toronto, eMARLIN-T-MM significantly reduces the total person delays by 58% to 74% across various bus occupancy levels compared to pre-timed signals, outperforming the other decentralized RL-based ATSCs. In addition, eMARLIN-T-MM can automatically adapt to changes in the levels of occupancy, allowing it to optimize the intersection performance in response to varying transit and traffic demands.
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 | 0.002 |
| 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 | 0.000 | 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".