Simulating Network Paths with Recurrent Buffering Units
DOI:
https://doi.org/10.1609/aaai.v37i6.25820Keywords:
ML: Applications, APP: Communication, APP: Web, ML: Deep Generative Models & Autoencoders, ML: Time-Series/Data StreamsAbstract
Simulating physical network paths (e.g., Internet) is a cornerstone research problem in the emerging sub-field of AI-for-networking. We seek a model that generates end-to-end packet delay values in response to the time-varying load offered by a sender, which is typically a function of the previously output delays. The problem setting is unique, and renders the state-of-the-art text and time-series generative models inapplicable or ineffective. We formulate an ML problem at the intersection of dynamical systems, sequential decision making, and time-series modeling. We propose a novel grey-box approach to network simulation that embeds the semantics of physical network path in a new RNN-style model called Recurrent Buffering Unit, providing the interpretability of standard network simulator tools, the power of neural models, the efficiency of SGD-based techniques for learning, and yielding promising results on synthetic and real-world network traces.Downloads
Published
2023-06-26
How to Cite
Anshumaan, D., Balasubramanian, S., Tiwari, S., Natarajan, N., Sellamanickam, S., & Padmanabhan, V. N. (2023). Simulating Network Paths with Recurrent Buffering Units. Proceedings of the AAAI Conference on Artificial Intelligence, 37(6), 6684-6692. https://doi.org/10.1609/aaai.v37i6.25820
Issue
Section
AAAI Technical Track on Machine Learning I