Methods for Creating Efficient Convolutional Neural Networks

In the past few years convolutional neural networks (CNN) have revolutionized several application domains in AI and computer vision. The biggest challenge with state-of-the-art CNNs is the massive compute demands that prevent these models from being used in many embedded systems and other resource-constrained environments. In this talk, I will explain and contrast several recent techniques that enable CNN models with high accuracy to consume very little memory and processor resources. These methods include a variety of algorithmic and optimization approaches to deep learning models. Quantization, sparsification and compact model design are three of the major techniques for efficient CNNs which will be discussed in the context of computer vision applications including detection, recognition and segmentation.