Disentangling Deep Neural Networks
Deep Neural Networks (DNN) can extract useful representations from unstructured data, which can be used for tasks like classification or object detection. In a joint publication in "Nature Nanotechnology", the team of Prof. Luca Benini at the Integrated Systems Laboratory (IIS) together with IBM Research Zürich present an efficient compute engine for disentangling these data-driven holographic representations.
Links
external pageRead article (IBM Research)
external pageIn-memory factorization of holographic perceptual representations (Nature Nanotechnology)
Integrated Systems Laboratory (IIS)