Search papers, labs, and topics across Lattice.
This paper introduces an adaptive retraining approach for Open Radio Access Networks (O-RAN) that utilizes Q-learning to optimize the balance between forecasting accuracy and retraining costs, framing the retraining decision as a Markov Decision Process (MDP). By employing a multi-expert Long Short-Term Memory (LSTM) ensemble, the method addresses catastrophic forgetting and enhances robustness to dynamic traffic variations. Experimental results demonstrate a significant reduction in retraining overhead compared to traditional greedy and random strategies, while ensuring compliance with Service Level Agreements (SLAs).
Reinforcement learning can drastically cut retraining costs in O-RAN without sacrificing performance, challenging traditional methods that rely on costly retraining.
Dynamic traffic variations in Open Radio Access Networks (O-RAN) lead to drift, which degrades the performance of Artificial Intelligence/Machine Learning (AI/ML) models. Traditional retraining approaches maintain forecasting accuracy but incur high computational cost and may lead to violations of Service Level Agreements (SLAs). This work proposes a Q-learning-based adaptive retraining approach that formulates the retraining decision as a Markov Decision Process (MDP), where a Reinforcement Learning (RL) agent learns a policy that balances forecasting accuracy and retraining cost. The proposed approach incorporates a multi-expert Long Short-Term Memory (LSTM) ensemble to mitigate catastrophic forgetting and improve robustness across diverse traffic conditions. Experimental results show that the proposed approach effectively reduces retraining overhead compared to greedy and random baselines, while maintaining system performance within predefined limits.
