ADR-X: ANN-Assisted Wireless Link Rate Adaptation for Compute-Constrained Embedded Gaming Devices
Offered By: USENIX via YouTube
Course Description
Overview
Explore a groundbreaking conference talk on ADR-X, an innovative ANN-based contextual multi-armed bandit rate adaptation technique for wireless gaming devices. Delve into the challenges of rapid channel variations in console gaming and learn how ADR-X achieves 10× lower packet losses than existing schemes while running 100× faster than state-of-the-art reinforcement learning approaches. Discover how this technique leverages communication theory and domain knowledge to create an efficient ANN suitable for deployment on embedded gaming devices. Gain insights into the extensive evaluations and user studies that revealed the limitations of current rate adaptation schemes and the need for significantly lower packet loss rates to ensure high-quality gaming experiences.
Syllabus
NSDI '24 - ADR-X: ANN-Assisted Wireless Link Rate Adaptation for Compute-Constrained Embedded...
Taught by
USENIX
Related Courses
Inside TensorFlow - TF-AgentsTensorFlow via YouTube Provably Efficient Reinforcement Learning with Linear Function Approximation - Chi Jin
Institute for Advanced Study via YouTube Exploration with Limited Memory - Streaming Algorithms for Coin Tossing, Noisy Comparisons, and Multi-Armed Bandits
Association for Computing Machinery (ACM) via YouTube How Slot Machines Are Advancing the State of the Art in Computer Go AI
Churchill CompSci Talks via YouTube CS885: Multi-Armed Bandits
Pascal Poupart via YouTube