Deep learning technology borrows from neural structure of brain. It grew in last decade. With growing fascination for data science and artificial intelligence , deep learning is a branch of machine learning to solve complex problems. It was first conceived after google cloud suite was born. Deep learning is a positive sign towards more data integrity and abstraction. Google brain is a project which deals with a machine level abstraction .
Supervised learning and unsupervised learning is a branch of science that learns from past mistakes. Backward ward propagation in neural network aids in machine leaning. It is based on the fact that learning is hierarchical meaning each layer’s learnings is achieved by auto correction . Neural network has many hidden layers and are subjected to weight correction. Deep learning has depth being added at very level. In image processing every depth gets rendered . Rina Detcher who first brought deep learning to public notice.
Deep learning has a variety of applications across platforms. Automatic speech recognition is most sought-after deep learning application. It renders voice to audio files. With advent of character recognition and image processing deep learning finds a place in biometrics. Neural language development is the branch that deals with human speech. Deep learning has contributed in human speed recognition. Minor areas like bioinformatics and image restoration has grown because of deep learning.
Deep learning is also susceptible to cyber-attacks. Cyber-attacks distorts deep learning algorithms . Information is extracted for personal use. There is also threat of data poisoning . Deep learning is dependent on algorithms . Platform scalability and platform independent has made deep learning todays feat . Convolutional algorithm and recurrent algorithm are important for ultimate implementation for projects.