Machine Learning

(making impossible possible)

Created by-
Vikrant Goswami EN 3rd Year(1602921173)
 Contents
• Introduction
• History
• Methods of Learning
• Applications
• Pros & Cons
• Conclusion
• References
 Introduction
• Field of study that gives computers the capability to learn
without being explicitly programmed.

• A branch of Artificial Intelligence

• Machine Learning
– Study of algorithms that
– improve their performance
– at some task
– with experience
 History
First NN in Deep Learning
1957 by came into AlphaGo beats
Alan Turing’s existence in 2006
Rosenblatt human in Go
‘Turing Test’

Arthur 1997 IBMs Google ,

Samuel’s ‘’Deep Blue’ Amazon, etc.
Checker increased
program research
Types of Learning
Supervised Learning:
 Supervised learning is where you have input variables (x) and an
output variable (Y) and you use an algorithm to learn the mapping
function from the input to the output

Unsupervised Learning:
 Unsupervised learning is where you only have input data (X) and
no corresponding output variables.
Reinforcement Learning:
 an area of machine learning concerned with how software agents
ought to take actions in an environment so as to maximize some
notion of cumulative reward
 Applications
• In Smart grid management
– Machine learning is being used in the management of smart grids by
predicting the energy consumption pattern of homes
– Strategic optimization at building level

• Load Forecasting
– Short term load forecasting is being done using ML algorithms for
quick and efficient forecasting(Machine learning based electric load
forecasting for short and long-term period,IEEE2018)
 Applications
 Autonomous Vehicles

 Taking over dangerous jobs

 Environmental Protection

 Enhanced healthcare

 Innovations in Banking

 Personalized digital media

 Pros & Cons
• Pros
– Easily identifies trends and patterns

– No human intervention needed (automation)

– Continuous Improvement

– Handling multi-dimensional and multi-variety data

– Wide Applications
 Pros & Cons
• Cons
– Data Acquisition

– Time and Resources

– Error Susceptibilty
 Conclusion
• Machine learning is providing a new and
better ways of doing things
• It has enhanced the standard of living
• Much harder job can be done much easily
• It is indeed a revolution for the betterment of
world
 References
• Distributed machine learning based smart-grid energy management with occupant
cognition, Hantao Huang, Hao Yu, Nanyang Technological University
• Mocanu, E. (2017). Machine learning applied to smart grids Eindhoven: Technische
Universiteit Eindhoven
• Machine Learning Technologies and Their Applications to Scientific and
Engineering Domains Workshop,August, 2016
• https://data-flair.training/blogs/advantages-and-disadvantages-of-machine-
learning/
• https://blog.valohai.com/smart-grids-use-machine-learning-to-forecast-load