Neural networks, one of the main techniques for deep learning, touch a number of different industries, performing such tasks as object recognition in vision applications, biometrics for security, and advanced driver assistance systems for autonomous vehicles.
In a 2016Embedded Vision Summit presentation by Dr. Chris Rowen, Chief Technology Officer for the IP Group at Cadence Design Systems titled “The Road Ahead for Neural Networks: Five Likely Surprises," the future of neural networks, including the fundamental capabilities and limitations of neural network computing, especially for real-time and embedded systems, are discussed. Rowen also talks about how the technology will likely evolve over the next decade, as well as the impact of neural networks on industry.
Additionally, he provides some forecasts on a series of developments, including unexpected changes in business models for data and training, in distribution of neural networks between the cloud and edge devices, in new types of hardware, and in novel software.
Check out Rowen’s entire presentation in this photo slideshow: