The superiority of humans over computers when it comes to recognizing objects or sounds is being explored by researchers at the Canadian Institute for Advanced Research (CIFAR) who want to develop computer algorithms to mimic the accuracy of our complex sensory systems. The Neural Computation and Adaptive Perception Program, which has members from the provinces of British Columbia, Ontario and Quebec, as well as from the US, Finland, Israel and the UK, wants to know how the human brain converts sensory stimuli into information and to recreate human-style learning in computers. Their research focuses on an artificial-intelligence technique called 'deep neural networks' to 'train' computers to recognize patterns, such as objects or a person's voice. Deep neural network concepts are practiced by companies including IBM, Microsoft and Google, and are dramatically reducing software error rates in applications from search engines to interactive voice programs.
展开▼
