filmov
tv
Using CFC Liquid Neural Nets to detect multiple values with the MyCaffe AI Platform
Показать описание
In this video we see how well the Liquid Closed-form continuous-time neural network (CfC) detects multiple values by detecting 25 values that follow the input values from a Sine curve. The tests shown in this video were implemented using the MyCaffe AI Platform and run with the MyCaffe Curve Gym.
CfC models were introduced by Ramin Hasani, et al., in the article ‘Closed-form continuous-time neural networks’ by Ramin Hasani, Mathias Lechner, Alexander Amini, Lucas Liebenwein, Aaron Ray, Max Tschaikowski, Gerald Teschl & Daniela Rus, 2022, nature machine intelligence.
CfC models were introduced by Ramin Hasani, et al., in the article ‘Closed-form continuous-time neural networks’ by Ramin Hasani, Mathias Lechner, Alexander Amini, Lucas Liebenwein, Aaron Ray, Max Tschaikowski, Gerald Teschl & Daniela Rus, 2022, nature machine intelligence.
0:01:51
0:02:33
0:00:59
0:02:00
0:13:00
0:49:30
0:00:33
0:10:24
0:10:39
0:45:28
0:19:40
0:05:48
0:37:41
0:41:10
0:00:49
0:01:54
0:22:52
0:00:45
0:31:26
0:01:01
0:01:01
0:03:18
0:01:21
0:04:17