Date of Publication :13th September 2017
Abstract: Deep learning, a subfield of machine learning concerned with algorithms is inspired or mimicked both by the structure, function and mechanism of brain and its neural structure called Artificial Neural Networks (ANN). It uses multiple layers of artificial neurons for classification and pattern recognition in its perception, analysis and understanding. Deep learning is a science that is aimed to produce algorithms mimicking and performing activities performed with human intelligence and perception associated with neural mechanisms without human intervention. In deep learning, we develop multiple layers of computations and analysis through algorithms that are embedded into chips as software, which results in better perception and results that are 90%- 100% accurate. This leads to a technology that is more informative, perceptive and efficient in a way, changing the world for better.
Reference :
-
- LeCun, Yann, Yoshua Bengio, and Geoffrey Hinton. "Deep learning." Nature 521.7553 (2015): 436- 444.
- Schmidhuber, Jürgen. "Deep learning in neural networks: An overview." Neural networks 61 (2015): 85- 117.