ADORN: Adaptive Drift handling for Open RAN using Reinforcement Learning
For O-RAN operators, it reduces computational cost of retraining without sacrificing performance, addressing a practical bottleneck in dynamic network environments.
ADORN uses Q-learning to adaptively decide when to retrain AI/ML models in O-RAN, reducing retraining overhead while maintaining forecasting accuracy within SLA limits.
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.