Liam Fedus
An AI researcher and co-founder of Periodic Labs, who previously led several large-scale model training projects at OpenAI.
Timeline (last 90 days)
A tweet appeared on X (formerly Twitter) that read: "Exciting results, @LiamFedus! Congrats to the whole team at Periodic Labs!"
Liam Fedus is one of OpenAI's core researchers and has participated in key work such as ChatGPT post-training and reinforcement learning.
Liam Fedus's research interests cover reinforcement learning, sparse models (such as the mixture-of-experts model MoE), and training efficiency issues of large-scale language models.
Liam Fedus previously led multiple large-scale model training projects at OpenAI, and Ekin Dogus Cubuk previously participated in the GNoME project at Google DeepMind.
All Facts (4)
Liam Fedus is one of OpenAI's core researchers and has participated in key work such as ChatGPT post-training and reinforcement learning.
50%UnverifiedLiam Fedus's research interests cover reinforcement learning, sparse models (such as the mixture-of-experts model MoE), and training efficiency issues of large-scale language models.
50%UnverifiedLiam Fedus previously led multiple large-scale model training projects at OpenAI, and Ekin Dogus Cubuk previously participated in the GNoME project at Google DeepMind.
50%UnverifiedA tweet appeared on X (formerly Twitter) that read: "Exciting results, @LiamFedus! Congrats to the whole team at Periodic Labs!"
50%