Glossary term
Glossary term
Infrastructure and Serving
A category of specialized hardware components designed to perform key computations needed for deep learning algorithms.
Accelerator chips (or just accelerators, for short) can significantly increase the speed and efficiency of training and inference tasks compared to a general-purpose CPU. They are ideal for training neural networks and similar computationally intensive tasks.
Examples of accelerator chips include:
Google's Tensor Processing Units (TPUs) with dedicated hardware for deep learning.
NVIDIA's GPUs which, though initially designed for graphics processing, are designed to enable parallel processing, which can significantly increase processing speed.
Examples of accelerator chips include:
Google's Tensor Processing Units (TPUs) with dedicated hardware for deep learning.
NVIDIA's GPUs which, though initially designed for graphics processing, are designed to enable parallel processing, which can significantly increase processing speed.
Created for this library
A genomics company uses GPU accelerator chips on Google Cloud to run protein folding inference fast enough to support interactive sessions for drug discovery teams.
A fintech firm provisions a fleet of TPU pods to retrain its anti-money-laundering model nightly across the full transaction history.
An autonomous trucking company deploys NVIDIA accelerators in-vehicle so the perception stack can process eight camera streams in real time at highway speeds.
Definition source: Google for Developers Machine Learning Glossary | Creative Commons Attribution 4.0 License