filmov
tv
TensorFlow on Modern Intel(R) Architectures
Показать описание
About the Session:
In this installment of the ALCF Many-Core Developer Sessions, AG Ramesh of Intel Corporation will lead a discussion on TensorFlow. First, researchers from the Theta Early Science Program will describe challenges their projects have encountered in this area. Following this, AG will share his extensive knowledge of performance optimization of machine learning on many-core architectures such as Knight's Landing. AG's slide presentation will be followed by an open Q&A.
About the Speaker:
AG Ramesh is a Sr. Software Engineer with Intel Corporation, where he works on machine learning and deep learning performance optimizations. He has been with Intel for over 20 years and has worked in a number of areas. Prior to his current role he developed computer vision-based applications using the Intel RealSense 3D camera technology. He also has extensive experience developing compiler and optimization tools. He has a Ph.D. in computer science from the State University of New York.
About the Series:
The ALCF Many-Core Developer Sessions are aimed at increasing the dialogue between ALCF users and developers of many-core systems and software. Attendees are encouraged to bring any questions they may have related to ALCF's Theta system and the Intel Xeon Phi technology in general.
In this installment of the ALCF Many-Core Developer Sessions, AG Ramesh of Intel Corporation will lead a discussion on TensorFlow. First, researchers from the Theta Early Science Program will describe challenges their projects have encountered in this area. Following this, AG will share his extensive knowledge of performance optimization of machine learning on many-core architectures such as Knight's Landing. AG's slide presentation will be followed by an open Q&A.
About the Speaker:
AG Ramesh is a Sr. Software Engineer with Intel Corporation, where he works on machine learning and deep learning performance optimizations. He has been with Intel for over 20 years and has worked in a number of areas. Prior to his current role he developed computer vision-based applications using the Intel RealSense 3D camera technology. He also has extensive experience developing compiler and optimization tools. He has a Ph.D. in computer science from the State University of New York.
About the Series:
The ALCF Many-Core Developer Sessions are aimed at increasing the dialogue between ALCF users and developers of many-core systems and software. Attendees are encouraged to bring any questions they may have related to ALCF's Theta system and the Intel Xeon Phi technology in general.
1:01:29
0:00:49
1:37:14
0:02:43
0:11:12
1:47:25
0:58:46
0:37:45
0:51:24
0:02:51
0:11:41
0:15:30
0:02:35
1:00:21
0:00:13
0:06:27
0:34:53
0:23:28
0:01:34
0:38:38
2:38:46
1:09:13
1:29:34