Work

Arithmetic-intensity-guided fault tolerance for neural network inference on GPUs

Year: 2021

Type: conference-paper

Abstract: Neural networks (NNs) are increasingly employed in safety-critical domains and in environments prone to unreliability (e.g., soft errors), such as on spacecraft. Therefore, it is critical to impart fault tolerance to NN inference. Algorithm-based fault tolerance (ABFT) is emerging as an efficient approach for fault tolerance in NNs.

Source: International Conference for High Performance Computing, Networking, Storage and Analysis (SC)

DOI: 10.1145/3458817.3476184

Authors: Jack Kosaian, K. V. Rashmi

Institution: Carnegie Mellon University

Language: English

Topic: Distributed systems and fault tolerance

Subfield: Computer Networks and Communications

Field: Computer Science

Domain: Physical Sciences

Open Access status: hybrid

Funders: National Science Foundation, Amazon Web Services, European Regional Development Fund

Awards: Graduate Research Fellowship Program (GRFP), Graduate Research Fellowship Program (GRFP), G274581676, G4975383095