Robust and Resource-efficient Machine Learning Aided Viewport Prediction in Virtual Reality

no code implementations • 20 Dec 2022 • Yuang Jiang, Konstantinos Poularakis, Diego Kiedanski, Sastry Kompella, Leandros Tassiulas

In this work, we propose a novel meta learning based viewport prediction paradigm to alleviate the worst prediction performance and ensure the robustness of viewport prediction.

Meta-Learning

State Action Separable Reinforcement Learning

no code implementations • 5 Jun 2020 • Ziyao Zhang, Liang Ma, Kin K. Leung, Konstantinos Poularakis, Mudhakar Srivatsa

We observe that although actions directly define the agents' behaviors, for many problems the next state after a state transition matters more than the action taken, in determining the return of such a state transition.

Decision Making reinforcement-learning +2

MACS: Deep Reinforcement Learning based SDN Controller Synchronization Policy Design

no code implementations • 19 Sep 2019 • Ziyao Zhang, Liang Ma, Konstantinos Poularakis, Kin K. Leung, Jeremy Tucker, Ananthram Swami

In distributed software-defined networks (SDN), multiple physical SDN controllers, each managing a network domain, are implemented to balance centralised control, scalability, and reliability requirements.

Deep Reinforcement Learning reinforcement-learning +1