Mixture-of-Supernets: Improving Weight-Sharing Supernet Training with Architecture-Routed Mixture-of-Experts

Ganesh Jawahar | Haichuan Yang | Yunyang Xiong | Zechun Liu | Dilin Wang | Fei Sun | Meng Li | Aasish Pappu | Barlas Oguz | Muhammad Abdul-Mageed | Laks Lakshmanan | Raghuraman Krishnamoorthi | Vikas Chandra |

Paper Details:

Month: August
Year: 2024
Location: Bangkok, Thailand and virtual meeting
Venue: F | i | n | d | i | n | g | s | - | A | C | L |

Citations

URL

No Citations Yet

Field Of Study