Liam Fedus
@LiamFedus
Building industrial-scale science at @periodiclabs
Past: VP of Post-Training @OpenAI; Google Brain
加入 October 2012
1.2K 正在關注 36.4K 粉絲
Small groups of people with relatively little money on the line (<$10k markets) produce well-calibrated predictions across diverse domains
We ran a study on over 2 million markets to evaluate their calibration. To our knowledge, this is the largest replicable study on this topic.
Here are the results: 🧵
