Day 51 – Q 1.What is deep learning? Explain. What are its applications? Discuss.
1. What is deep learning? Explain. What are its applications? Discuss.
डीप लर्निंग क्या है? समझाएं। इसके अनुप्रयोग क्या हैं? चर्चा करें।
A Fourth Industrial Revolution is building as digital revolution that has been occurring since the middle of the last century. It is characterized by a fusion of technologies that is blurring the lines between the physical, digital, and biological spheres. Artificial intelligence is a crucial part of this fourth industrial revolution. And Deep learning is one of the important aspects of Artificial Intelligence.
Deep Learning is a subfield of machine learning concerned with algorithms inspired by the structure and function of the brain called artificial neural networks.
Deep learning (aka deep structured learning or hierarchical learning) is part of a broader family of machine learning methods based on learning data representations, as opposed to task-specific algorithms.
Deep learning neural network consists of two terms:
1) Neural networks, a biologically-inspired programming paradigm which enables a computer to learn from observational data.
2) Deep learning, a powerful set of techniques for learning in neural networks
Application of Deep learning
- Self-driving cars where the neural networks can be trained over parameters such as traffic patterns, traffic rules, weather and road quality etc. and it can self-improvise to drive efficiently.
- Weather forecasting where training parameters could be wind pattern, air-pressure, temperature and previous weather records of the year etc. so that it could predict weather phenomena without human intervention.
- Automatic machine translation, deciphering complicated scripts and language modeling.
- Automatic Game Playing – A recent example is AlphaGo which beat the world champion.
- Examination of huge amount of space data to come out with patterns and new discoveries.
- Robotics and Internet of Things (IoT) are the areas that can significantly improve our interaction with outer world.
- Restoration of old paintings, identification through low resolution images, automatic music composing etc.
- Border management with hostile neighbourhood can be done through computer vision, including object recognition, etc. with minimal loss of life and property.
- Crime and investigations, huge pending cases are serious problem before Indian judiciary and administration. Machine learning can help in solving cases, bioinformatics and DNA profiling etc.
- There is huge potential for machine learning in realising the dream of new India but this will not happen until the personal biases mentioned previously, and these “greed” biases are addressed.
- It is also not recommended to completely rely on machines in certain sectors such as health and such other areas of human interface.
- It must be taken care that replacement of human with intelligent machines would not increase the socio-economic inequalities and deprivations in society