From automatic differentiation to message passing
4 914
14.5
Microsoft Research334 тыс
Следующее
06.08.19 – 1 0961:06:43Популярные
58 дней – 68354:47
258 дней – 1 0285:33Опубликовано 6 августа 2019, 16:00
See updated video here: microsoft.com/en-us/research/v...
Automatic differentiation is an elegant technique for converting a computable function expressed as a program into a derivative-computing program with similar time complexity. It does not execute the original program as a black-box, nor does it expand the program into a mathematical formula, both of which would be counter-productive. By generalizing this technique, you can produce efficient algorithms for constraint satisfaction, optimization, and Bayesian inference on models specified as programs. This approach can be broadly described as compiling into a message-passing program.
Automatic differentiation is an elegant technique for converting a computable function expressed as a program into a derivative-computing program with similar time complexity. It does not execute the original program as a black-box, nor does it expand the program into a mathematical formula, both of which would be counter-productive. By generalizing this technique, you can produce efficient algorithms for constraint satisfaction, optimization, and Bayesian inference on models specified as programs. This approach can be broadly described as compiling into a message-passing program.
Свежие видео
3 дня – 29114:09
4 дня – 5 0240:31
13 дней – 3 74511:36
16 дней – 397 2944:25
325 дней – 2 0808:03Случайные видео
12.10.23 – 2990:15
24.05.23 – 426 8190:28
27.03.23 – 19 8080:56
03.10.22 – 158 2900:37
26.03.22 – 5457:23
31 день – 14 6132:01
3 дня – 62253:30
3 дня – 019:31
3 дня – 40 81927:39
7 дней – 1668:50
9 дней – 7310:32
11 дней – 2 4033:06
13 дней – 1 4150:21
