Explainable AI Whiteboard Technical Series: Reinforcement Learning

Skip ad
37 Views

  • Thanks! Share it with your friends!

    You disliked this video. Thanks for the feedback!

juniper
Published
The explainable AI whiteboard technical series explores and explains some of the key tools within our data science toolbox that powers the Juniper AI-Native Networking Platform. In this video we cover reinforcement learning.
Radio frequency (RF) environments are inherently complex, making it difficult to optimize data transmission. Radio Resource Management (RRM) has long been used to manage RF in wireless LANs, but manual approaches, such as site surveys, cannot adapt to changing environments with people, devices, and external interferences like microwaves or radar.

Juniper's AI-driven Mist wireless solution uses reinforcement learning to automate RRM, adjusting to both interference and movement in real-time. Reinforcement learning allows the system to improve by trial and error, optimizing three factors: coverage, capacity, and connectivity. It adjusts settings like band selection, transmit power, channels, bandwidth, and BSS color for optimal network performance.

The machine’s value function determines which actions yield the best wireless experience. Actions, such as increasing transmit power or switching bands, build a policy that maximizes future rewards, ensuring optimal network service even as environments change. As time goes on, the network continually learns and adjusts, offering a tailored, real-time wireless environment for each site.

Mist’s AI-driven approach ensures that even large organizations with multiple sites maintain high-quality, adaptable wireless coverage, addressing site-specific variances and delivering custom wireless environments in real-time.

Chapters:
0:00: Introduction
0:17: Challenges
2:55 Benefits
Category
Juniper Networks
Tags
AI, ML, XAI
Be the first to comment