eBandit: Kernel-Driven Reinforcement Learning for Adaptive Video Streaming
2026-04-09 • Networking and Internet Architecture
Networking and Internet ArchitectureArtificial Intelligence
AI summaryⓘ
The authors created eBandit, a system that moves video quality control inside the operating system kernel, allowing it to see important network details usually hidden from user-level programs. Using a simple learning method called a Multi-Armed Bandit, eBandit tests different quality settings based on live internet connection data and picks the best one quickly. This approach performs better than traditional fixed methods in both simulated and real mobile network conditions, improving overall video experience. Their work shows that putting decision-making closer to the network can lead to smarter video streaming.
Adaptive Bitrate (ABR)Linux kerneleBPFMulti-Armed BanditQuality of Experience (QoE)TCP metricssockopsepsilon-greedyRTTtransport layer
Authors
Mahdi Alizadeh
Abstract
User-space Adaptive Bitrate (ABR) algorithms cannot see the transport layer signals that matter most, such as minimum RTT and instantaneous delivery rate, and they respond to network changes only after damage has already propagated to the playout buffer. We present eBandit, a framework that relocates both network monitoring and ABR algorithm selection into the Linux kernel using eBPF. A lightweight epsilon-greedy Multi-Armed Bandit (MAB) runs inside a sockops program, evaluating three ABR heuristics against a reward derived from live TCP metrics. On an adversarial synthetic trace eBandit achieves $416.3 \pm 4.9$ cumulative QoE, outperforming the best static heuristic by $7.2\%$. On 42 real-world sessions eBandit achieves a mean QoE per chunk of $1.241$, the highest across all policies, demonstrating that kernel-resident bandit learning transfers to heterogeneous mobile conditions.