Context Detection and Identification In Multi-Agent Reinforcement Learning With Non-Stationary Environment
Bibliographic record
Abstract
Reinforcement learning methods are mostly constructed on the very assumption that environments are stationary. However, most real world environments are non-stationary; that is, we assume they are composed of several stationary components (i.e., sub-environments or contexts). So, methods with this assumption are not capable of learning non-stationary environments. Reinforcement Learning - Context Detection (RL-CD) method enables the agent to learn the environment without prior information; detect the environment’s context change points and create a partial model for each context. The underlying environment of this approach is single-agent and has shortcomings for multi-agent learning. In this study, we introduce a new approach called Multi-agent reinforcement learning-context detection (MARL-CD), which can both detect context change points and enable agents to learn non-stationary environments with multi-agent settings. This approach is based on RL-CD approach. MARL-CD is more efficient in terms of detecting context change created by the agents on the environment and detecting the context change of the environment itself. It enables an agent to detect the context changes not only from the change of environment dynamics but also from policy changes of agents in the environment. In the approach in this study, it has been shown by the experimental results that the agents spend 16% less energy and are more efficient than RL-CD in terms of detecting the change points more accurately.
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 | 0.004 |
| 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 | 0.001 | 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".