Multiagent Deep-Reinforcement-Learning-Based Virtual Resource Allocation Through Network Function Virtualization in Internet of Things
Bibliographic record
Abstract
Resource allocation is a significant task in the emerging area of Internet of Things (IoT). IoT devices are usually low-cost devices with limited computational power and capabilities for long term communication. In this article, the network function virtualization (NFV) technique is used to access resources of the network and a reinforcement learning (RL) algorithm is used to solve the problem of resource allocation in IoT networks. The traffic of the IoT network uses the substrate network which is available through NFV for its data transmission. The data transmission needs of the IoT network are translated to virtual requests and service function chain (SFC) are mapped to the substrate network to serve the requests. The problem of SFC placement while meeting the system constraints of the IoT network is a nonconvex problem. In the proposed deep RL (DRL)-based resource allocation, the virtual layer acts as a common repository of the network resources. The optimization problem of SFC placement under the system constraints of IoT networks can be formulated as a Markovian decision process (MDP). The MDP problem is solved through a multiagent DRL algorithm where each agent serves an SFC. Two Q-networks are considered, where one Q-network solves the SFC placement problem while the other updates weights of the Q-network through keeping track of long-term policy changes. The virtual agents serving SFCs interact with the environment, receive reward collectively and update the policy by using the learned experiences. We show that the proposed scheme can solve the optimization problem of SFC placement through adequate reward design, state, and action space formulation. Simulation results demonstrate that the multiagent DRL scheme outperforms the reference schemes in terms of utility gained as measured through different network parameters.
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 | 0.001 |
| 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".