A new hope for network model generalization
This work addresses the challenge of model generalization in networking, which is considered a lost cause, offering a potential solution for researchers and practitioners in the field, though it is incremental as it adapts an existing architecture.
The paper tackles the problem of generalizing machine learning models for network traffic dynamics by proposing a Network Traffic Transformer (NTT), adapted from transformers, which shows promising initial results in generalizing to new prediction tasks and environments.
Generalizing machine learning (ML) models for network traffic dynamics tends to be considered a lost cause. Hence for every new task, we design new models and train them on model-specific datasets closely mimicking the deployment environments. Yet, an ML architecture called_Transformer_ has enabled previously unimaginable generalization in other domains. Nowadays, one can download a model pre-trained on massive datasets and only fine-tune it for a specific task and context with comparatively little time and data. These fine-tuned models are now state-of-the-art for many benchmarks. We believe this progress could translate to networking and propose a Network Traffic Transformer (NTT), a transformer adapted to learn network dynamics from packet traces. Our initial results are promising: NTT seems able to generalize to new prediction tasks and environments. This study suggests there is still hope for generalization, though it calls for a lot of future research.