Deep Learning Glossary

The Automated Neural Architecture Construction (AutoNAC) technology is an algorithmic optimization engine that squeezes maximum utilization out of any hardware. It contains a Neural Architecture Search (NAS) component that redesigns...

AutoML (Automated Machine Learning) makes machine learning available and accessible by automating machine learning. It automates the standard (manual) machine learning process with off-the-shelf methods. AutoML shortens the process of pre-processing...

When you use a model ‘off the shelf,’ it generally comes with a suggested training recipe. The thing is, these models are usually trained on very powerful GPUs, which may...

In deep learning, a convolutional neural network (CNN) is a type of artificial neural network commonly used for processing and analyzing visual imagery recognition. Compared with traditional neural networks, CNNs...

Data-aware model development is an approach to deep learning development that takes into account the training data, along with the baseline model and target inference hardware. Using AutoNAC, the final...

Deep learning inference is the phase in development where the capabilities learned during training is put to work. The trained deep neural networks (DNN) make predictions (or inferences) on new...

Deep learning inference acceleration is the end-to-end process of accelerating the inference of neural models while preserving the baseline accuracy. It is fully aware of the desired target inference hardware...

In deep learning development, the inference stack is a set of components that influence inference computation and performance. Also called the acceleration stack, it can be divided into three different...

Deep Neural Network (DNN), also called Deep Nets, is a neural network with a certain level of complexity. It can be considered as stacked neural networks, or networks composed of...

The deep learning efficient frontier is the curve that determines what is the best accuracy one can achieve given a specific latency, or vice versa. In other words, it is...