Type of company
Select company type
Select company status preset
Number of employees
London, United Kingdom
Artificial Learning Ltd is working with leading UK universities to embody proven machine learning algorithms in ASICs (application-specific integrated circuits). Our focus is on developing ultra-efficient, highly-scalable devices for machine learning tasks in mobile autonomous systems. Our business strategy comprises disruptive innovation through synthesis of titan, cutting-edge advances in machine-learning theory and VLSI technology. Our market strategy made the most-voted list front page in Technology Entrepreneurship 2013 at Stanford University. Follow us with the buttons to the right. To see everyone involved in our venture, please see our research group web pages. Established 2012. Company No. 08168963 registered in England & Wales. We comply with the UK Data Protection Act.
Machine Learning on a Chip | Artificial Learning
... Restricted Boltzmann Machines and Deep Belief Networks ...
New Delhi, India
FaceChk as the name suggests is an intelligence system that can read and identify Face & retina. We identified that Face, Speed and accuracy are the future and our system gives all these. FaceChk is an organization working towards a contactless (or touchless) system. Facechk’s unique face recognition technology leverages deep learning & AI incorporating several custom algorithms to provide high accuracy and portability across a variety of inexpensive devices such as low-cost cameras, smartphones & NPU (Neural Processing Unit) devices. The technology has been benchmarked with the world’s best and has proven to be one of the most accurate in a global open challenge.
Technology - Face Recognition Solutions :: Contactless Biometrics System
... TECHNOLOGY Heart of FaceChk Deep Learning Inspired by nervous systems DNN, CNN, DBN, RNN, LSTM Semi, Fully, Un-Supervised Deep Belief Networks Deep Boltzman Machines Micro- ...
A Deep Belief Network (DBN) is a type of artificial neural network consisting of multiple layers of restricted Boltzmann machines (RBMs) that are connected to form a deep network. It is trained using a greedy layer-wise unsupervised learning algorithm, with each layer learning to represent different levels of abstraction in the data. DBNs can be used for classification and regression tasks, as well as for feature extraction.