Get our free extension to see links to code for papers anywhere online!

Add to Chrome

Add to Firefox

Picture for Sameer Kumar

Sameer Kumar

Designing Effective Sparse Expert Models

Feb 17, 2022
Barret Zoph, Irwan Bello, Sameer Kumar, Nan Du, Yanping Huang, Jeff Dean, Noam Shazeer, William Fedus

Figure 1 for Designing Effective Sparse Expert Models

Figure 2 for Designing Effective Sparse Expert Models

Figure 3 for Designing Effective Sparse Expert Models

Figure 4 for Designing Effective Sparse Expert Models

* 25 pages main text, 39 pages overall

Access Paper or Ask Questions

Highly Available Data Parallel ML training on Mesh Networks

Nov 06, 2020
Sameer Kumar, Norm Jouppi

Figure 1 for Highly Available Data Parallel ML training on Mesh Networks

Figure 2 for Highly Available Data Parallel ML training on Mesh Networks

Figure 3 for Highly Available Data Parallel ML training on Mesh Networks

Figure 4 for Highly Available Data Parallel ML training on Mesh Networks

Access Paper or Ask Questions

Image Classification at Supercomputer Scale

Dec 02, 2018
Chris Ying, Sameer Kumar, Dehao Chen, Tao Wang, Youlong Cheng

Figure 1 for Image Classification at Supercomputer Scale

Figure 2 for Image Classification at Supercomputer Scale

Figure 3 for Image Classification at Supercomputer Scale

Figure 4 for Image Classification at Supercomputer Scale

* Presented as part of Systems for ML Workshop @ NIPS 2018

Access Paper or Ask Questions