Slicing for AI: An Online Learning Framework for Network Slicing Supporting AI Services

TL;DR

This paper proposes an online learning framework to determine the allocation of computational and communication resources to AI services, to optimize their accuracy as one of their unique key performance indicators (KPIs) while abiding by resources, learning latency, and cost constraints.

Abstract

The forthcoming 6G networks will embrace a new realm of AI-driven services that requires innovative network slicing strategies, namely slicing for AI, which involves the creation of customized network slices to meet Quality of service (QoS) requirements of diverse AI services. This poses challenges due to time-varying dynamics of users' behavior and mobile networks. Thus, this paper proposes an online learning framework to optimize the allocation of computational and communication resources to AI services, while considering their unique key performance indicators (KPIs), such as accuracy, latency, and cost. We define a problem of optimizing the total accuracy while balancing conflicting KPIs, prove its NP-hardness, and propose an online learning framework for solving it in dynamic environments. We present a basic online solution and two variations employing a pre-learning elimination method for reducing the decision space to expedite the learning. Furthermore, we propose a biased decision space subset selection by incorporating prior knowledge to enhance the learning speed without compromising performance and present two alternatives of handling the selected subset. Our results depict the efficiency of the proposed solutions in converging to the optimal decisions, while reducing decision space and improving time complexity.

Figures (8)

• Figure 1: System model under study.
• Figure 2: The online learning framework, representing the interaction between the online optimization controller and a performance monitoring module
• Figure 3: A representation of different initial probability distribution schemes
• Figure 4: a) Cumulative regret curve using different values for $\eta$ compared against the regret bound at $\eta_\text{op}$, b) Average reward curve of OLS against the benchmarks using the best observable $\eta$, c) Average regret curve considering different numbers of DL model deployment requests using the best observed $\eta$
• Figure 5: Comparison between the three elimination methods: a) Average regret curve with $\eta = 0.001$, b) Action probability curve showing the probability given to the optimal action with $\eta = 0.001$ , c) Average regret curve using the best observable $\eta$, c) Action probability curve showing the probability given to the optimal action using the best observable $\eta$.