Abstract:Reinforcement learning (RL) is currently used in various real-life applications. RL-based solutions have the potential to generically address problems, including the ones that are difficult to solve with heuristics and meta-heuristics and, in addition, the set of problems and issues where some intelligent or cognitive approach is required. However, reinforcement learning agents require a not straightforward design and have important design issues. RL agent design issues include the target problem modeling, state-space explosion, the training process, and agent efficiency. Research currently addresses these issues aiming to foster RL dissemination. A BAM model, in summary, allocates and shares resources with users. There are three basic BAM models and several hybrids that differ in how they allocate and share resources among users. This paper addresses the issue of an RL agent design and efficiency. The RL agent's objective is to allocate and share resources among users. The paper investigates how a BAM model can contribute to the RL agent design and efficiency. The AllocTC-Sharing (ATCS) model is analytically described and simulated to evaluate how it mimics the RL agent operation and how the ATCS can offload computational tasks from the RL agent. The essential argument researched is whether algorithms integrated with the RL agent design and operation have the potential to facilitate agent design and optimize its execution. The ATCS analytical model and simulation presented demonstrate that a BAM model offloads agent tasks and assists the agent's design and optimization.

G-BAM: A Generalized Bandwidth Allocation Model for IP/MPLS/DS-TE Networks

A SDN/OpenFlow Framework for Dynamic Resource Allocation based on Bandwidth Allocation Model

Evaluating the Applicability of Bandwidth Allocation Models for EON Slot Allocation

Bandwidth Resource Dynamic Allocation in Differentiated Services Model

Bandwidth-based Application-Aware Multipath Routing for NoCs

Reinforcement Learning Agent Design and Optimization with Bandwidth Allocation Model

Joint Bandwidth Scheduling to Support Differentiated Services and Multiple Service Providers in 1G and 10G EPONs

A Fair Dynamic Bandwidth Allocation Algorithm for EPON with Service Level Agreement Supporting

Eba: Efficient Bandwidth Guarantee under Traffic Variability in Datacenters

Multi-granularity Bandwidth Allocation for Large-Scale WDM/TDM PON

A Dynamic Bandwidth Allocation Scheme for EPON

A Generalized Satisfaction Oriented Bandwidth Management

Modeling and Accomplishing the BEREC Network Neutrality Policy

A Priority-Based Processor Sharing Model for TDM Passive Optical Networks

Dynamic Bandwidth Reservation Admission Control Scheme for the IEEE 802.16e Broadband Wireless Access Systems.

Dynamic Bandwidth Allocation for C-Band Shared FBG Sensing and Telecommunications

Profit oriented multichannel resource management for integrated Internet and DVB-T network

A Novel Dynamic Bandwidth Allocation Scheme Based on International Standard in Ethernet PON

A dynamic routing and wavelength assignment algorithm in IP/MPLS over WDM networks

A Novel Upstream Dynamic Bandwidth Assignment Scheme for GPONs

Bandwidth Reallocation for Bandwidth Asymmetry Wireless Networks Based on Distributed Multiservice Admission Control.