Machine Learning

 welcome to machine learning for

engineering and science applications

I am Balaji Srinivasan I am in the

mechanical engineering department

hi I'm gonna put the Krishna mood I am

in the department of engineering design

and both of us are from iit madras so if

we look at various applications that all

of us are using already in real life for

example this is Amazon's recommender

system some of you might have seen

Amazon echo which is a speech

recognition system of course everybody

has used Gmail spam classifier and this

is Google Lexus this is a latest

self-driving car

all of these use machine learning

algorithms in one way or the other

our purpose in this course is to try and

utilize the same algorithms for more

general problems for example medical

image diagnosis or for speeding up CFD

computations we look at the course aims

basically we will try to understand some

of the basic machine normal learning

models thoroughly with specific emphasis

on deep learning which is the current

state-of-the-art in machine learning for

us since recent advances in cans or

generated vessel networks and also in

reinforcement learning we will learn to

apply these techniques to problems in

engineering for example problems in

medical image analysis as well as

turbulence modeling in computational

fluid dynamics we expect you to learn to

program in Python and also to learn to

program in popular deep learning

platforms like packet arch and tensor

flow will feel the course auckland the

course consists of three broad parts the

part one will be focused on artificial

neural networks and deep learning or

deep neural networks which will include

CN NS and or enhance or recurrent neural

networks and other T will also cover

other classical techniques like binary

decision tree Sandom forest and

probabilistic techniques we will

conclude the course with with a look at

the recent advances in the fields of

deep learning which includes

variational autoencoders auto-encoders

in general generative models which

include generality of additional

networks as well as a small introduction

to reinforcement learning

thank you