Customized Video QoE Estimation with Algorithm-Agnostic Transfer Learning

Abstract

The development of QoE models by means of Machine Learning (ML) is challenging, amongst others due to small-size datasets, lack of diversity in user profiles in the source domain, and too much diversity in the target domains of QoE models. Furthermore, datasets can be hard to share between research entities, as the machine learning models and the collected user data from the user studies may be IPR-or GDPRsensitive. This makes a decentralized learning-based framework appealing for sharing and aggregating learned knowledge inbetween the local models that map the obtained metrics to the user QoE, such as Mean Opinion Scores (MOS). In this paper, we present a transfer learning-based ML model training approach, which allows decentralized local models to share generic indicators on MOS to learn a generic base model, and then customize the generic base model further using additional features that are unique to those specific localized (and potentially sensitive) QoE nodes. We show that the proposed approach is agnostic to specific ML algorithms, stacked upon each other, as it does not necessitate the collaborating localized nodes to run the same ML algorithm. Our reproducible results reveal the advantages of stacking various generic and specific models with corresponding weight factors. Moreover, we identify the optimal combination of algorithms and weight factors for the corresponding localized QoE nodes.

Real-Time Classification of Real-Time Communications

“…Instead, the approach proposed in [34] used it in combination with Deep Reinforcement Learning to solve the reconfiguration problem in the context of experience-driven networking. It has also been used for QoE estimation of video streaming [35], [36]. The transfer learning technique we use (CORAL [26]) has already been used in optical networks for assisted quality of transmission estimation of an optical lightpath [37].…”

On Machine Learning Based Video QoE Estimation Across Different Networks

“…Applying the model trained on Amazon, YouTube, and Twitch data to Netflix data resulted in a significant drop in model performance. Regarding the generalization efforts, interesting approaches can be found in [31], [32], where the authors investigate challenges related to model sharing and a transfer-learning approach which allows local models to learn a generic base model for MOS, and then consider additional features for location-specific QoE models. However, both approaches rely on application-level KPIs and do not consider estimating QoE from encrypted network traffic.…”