Modifying Neural Networks in Adversarial Agents of Multi-agent Reinforcement Learning Systems
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
This paper proposes a method to reduce the malicious agent’s negative effects on a multi-agent reinforcement learning (MARL) system, including actor-critic architecture. The method achieves the overall goal of the MARL system, which is to increase the cumulative reward of all individual agents and reduce the malicious agents’ harmful effects on the entire MARL system. Assuming that the adverse agent is detectable, we propose to change the malicious agent’s neural network (NN) structure. By leveraging a comparative methodology, we have demonstrated that a specific NN architecture using a linear activation function surpasses another utilizing a sigmoid activation function in minimizing loss. Our analysis indicates that this performance differential is attributable to the utilization of distinct activation functions within the models. This approach involves calculating the gradient of the loss function with respect to the activation function. The claims have been proven theoretically, and the simulation confirms theoretical findings.
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.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 it