Introduction to Artificial Intelligence - Computer Science & Engineering Video Lecture - Computer Science Engineering (CSE)

Oh
welcome to this course so we are going
to do this course in a slightly
different way this time because of the
audiovisual way in which we are going to
do this possibly for many of you this
might be a new kind of experience I am
doing it for the second time so it is
quite new for me as well I like to teach
more on blackboard but let us try this
out for once instead and see how it goes
so the subject that we are going to
cover in this course is artificial
intelligence and this is one subject
which has which is really old it has
been there for a long long time every
subject has up and down there is some
point of time where there is a lot of
hype about the subject so in AI also
there was a time when people claimed a
lot and yet at that point of time
computation was not up to that mark to
really make all that happen so that hype
has gone down now and we now know better
that what we can do and what we cannot
do so this course is to show you some of
the things that we can do and also to
show you what are the things that AI
promises for the future and because the
hype is not there so much now so we can
really go down to what actually are the
feasible things because we understand
computation a lot better now than maybe
70 years before so in this course the
course number is CS 4 0 0 2 and so I
will start with describing what is
called the Turing test
now this stiring test was proposed by
Alan curing in 1950 and the idea is like
this that you have a computer and it is
being interrogated by a human why a
teletype so what happens is that you
have a partition
and you have a computer on one side and
you have a human being sitting on the
other and the human being interrogate
the computer it does not know whether on
the other side there is a computer or
there is a human being and if the
computer can give the answers in such a
way that the human being is not able to
determine whether there is a computer on
the other side or a human being then
this computer passes the Turing test now
this test was proposed in 1950 at that
time we had hardly any of natural
language processing we did not have much
of logic and deduction so it was quite a
tall ask to really expect a computer to
be able to do this but even at that time
there were people who challenged this
very idea
they said that just because a computer
is able to answer in a human-like
fashion to deceive a human being does
not really mean that artificial
intelligence is born and so they came up
with what is called the Chinese room
argument and the Chinese room argument
was something like this that you have a
room
inside the room there is an operator who
does not know Chinese but that operator
does have a huge volume of Chinese
literature so what does it have so here
there is some Chinese people who are
sending in some Chinese texts and what
this person does is it just looks up
this table this document and whatever is
written on the other side corresponding
to this text it simply outputs that and
the person on this side believes that
the person who is inside knows Chinese
because it is being able to read in
Chinese and again produce something
gained in Chinese but actually this
person here does not know Chinese so it
is just a matter of translating so what
was conjectured at that time is there is
a lot of translations automatic
translations that a computer should be
able to do and if we are able to do that
smartly enough then it is not impossible
to be able to deceive a person but that
is not that does not mean that the
semantic understanding of the language
is there with the individual so even
today when you have a computer
translating from one language to the
other it does not necessarily mean that
the computer is able to semantically
understand the language so cognition and
understanding is a different thing as
compared to simple translation from one
language to other so let us try to see
what are the basic things around which
AI has been built the first thing that
people tried to do was to address the
problem of automated problem solving so
people try to see that how much can the
computer aiders in our ability to solve
problems
so the first kinds of things that people
came up in automated problem solving was
search
and search is nothing but an efficient
trial and error kind of algorithm so
what happens is I do not know actually
how to solve the problem the problem is
there I do not know the exact steps in
solving the problem
take for example sorting today I know
exactly I know five different algorithms
for sorting right if suppose you are the
first person who has come up with a
sorting problem then at that time you do
not know how to best write a sorting
algorithm so what we do is in that case
we adopt a trial and error method we try
something see if we reach the solution
otherwise we try something else
and reach the solution so very nice way
of sorting would be that you just
randomly reorder the elements and see
whether they are sorted or not right so
search starts from there and then when
you have problems which are
combinatorially complex and there is no
deterministic polynomial time algorithm
for solving it then we adopt a trial and
error search techniques so what happens
is the kind of problems that we will
address in this course will have
enormous computational complexity so if
you remember from your algorithms course
that there are problems which are that
can be solved in polynomial time others
which are np-complete or np-hard and
still harder problems and then in search
what happens is we come across what is
called a space-time trade-offs so if you
have more space then you are able to
solve a problem better if you have less
space then here come the time complexity
goes up so we will study in our search
algorithms at how we can balance out
space and time for really big problems
and then we will study what is called
heuristics that is the use of
domain-specific knowledge to accelerate
the search process
there has been several paradigms of
automated problem solving which has
evolved over the past some of them are
as follows we have linear programming
linear programming for a long time was
not known to be in P then do you know
who finally showed that it is in P there
is a karma curse algorithm Karmakar who
showed that linear programming is in P
so that was the one of the major
breakthroughs in computation then
integer programming see these are some
of the topics which we do not study as
part of the CS course but we should
study as part of the CS course because
these are also very well-defined and
very efficient automated problem solving
mechanisms integer programming basically
addresses problems which are harder than
linear programming and if time permits
we may have have a peep into this you
have already studied dynamic programming
as part of your algorithms course this
is also useful for solving large
communal combinatorial problems which
satisfy a principle of optimality we
will discuss that later not all problems
are amenable to be solved with dynamic
programming but some of them have a
specific property which makes them
useful to be solved by dynamic
programming techniques then we will
study a subject called heuristic search
which is the first topic that we will
study after this lecture where we will
see how we can use domain-specific
knowledge to aid this and we will also
study a little bit of evolutionary
algorithms which includes things like
neural networks and genetic algorithms
which have come up more recently and are
a slightly different kind of search
algorithms now the enormous problem
solving complexity rendered most of
these paradigms to be applicable to very
limited domains
when they were born and then what
happened was during 1985 to 1995
computation became free so if you go
back to some time before 1985 that is
about that time when we were in high
school and then we join the IIT for
doing our BTech so at that at those
times you know people used to queue up
in front of the computer center from
five o'clock in the morning with a pack
of punch cards right and those were
punch cards which contain Fortran
programs so people used to go and run
them and then if you had printed d0
instead of do in Fortran then it means
that you queue up again next morning at
five o'clock right and then when we join
that the systems were not booting by
themselves so you have to actually there
was a boot sequence you have to there
were no micro processors at that time so
there were the the CPU was a PCB which
had a separate ALU and separate
components and you actually had to
initialize all the registers and set the
program counter and then set the
computer running and we used to crave
for learning how to boot a computer
right so from that to sometime if you
look at in 1995 there is an enormous
change in the amount of computation
power that you have on the desktop so
today on on your desktop you have more
power than you had on the best computers
on that day so this has helped us in
really being able to solve larger
problems and AI is becoming a reality
through this process but there are other
bottlenecks which I will come to the
second thing that we want to do in AI is
to be able to use knowledge and to have
the ability of deduction
so here the fundamental question is how
to store and retrieve knowledge and then
how to interpret facts and rules and be
able to deduce
now there is a gap between what we call
knowledge and what we mean by
realization or understanding so I give a
example for this that imagine you are
asking a baby to learn colors see color
is a concept when you tell a baby that
look this ball is red it does not know
whether it is red because of the shape
because of the weight or because of the
dimensions it does not know what color
is so when you say that this is that it
does not know why this is that maybe it
will pick up another round thing and say
that is red too right so there is a
point of time when parents start getting
frustrated and they say that my child is
never going to learn colors and then one
fine day the child realizes that what is
the concept behind colors and on that
day it will start sighting out all kinds
of colors and say yes this is at this
green this is black right so what I am
saying is that you can impact knowledge
by saying that this is red this is blue
but then as long as you have that
knowledge it does not mean that you have
understood the concept called colors so
similarly when we are talking about
deduction you can have lots of data in
your hard disk you can have lots of data
in your database but that does not mean
that you have actually learned the rules
which governed the domain in which you
are region right so there is a gap
between this knowing how all that is
there to know and actually understanding
the rules and being able to interpret
and use those rules to deduce new things
so we will see what kinds of logic that
enable us to do this so we will study
first-order predicate logic we will
study propositional logic and a little
bit of other kinds of logic to see how
we can do reasoning with those kinds of
logics then there is another interesting
subject which has come up in the past
decade it's on the logics of knowledge
now this is a very interesting concept
this concept is I am just giving you a
glimpse of certain things that are
interesting and that may be covered in
this course so a logic of knowledge is
like this that suppose we take two young
children who have come home after play
so they have mud on their forehead now
their mother tells the children that
whoever has mud on his forehead just go
and wash now initially both wait they do
not go and then they go together let us
understand why see when they wait
actually one is seeing whether the other
is going because they do not know
whether they have mud on their own
forehead so they the person a assumes
that okay person B has mud on his
forehead so he must be going for washing
but then he sees that B is not going
then Y is B not going because B has seen
someone else and namely me that I have
mud on my forehead and just because I
was waiting for a very similar reason he
is also waiting for me to go so in the
second cycle they both realize that they
have mud on their respective foreheads
and they then they go for partial so
there is again a difference between me
knowing you knowing that I know and I
knowing that you know that I know right
so this is the kind of knowledge
hierarchy that will build up and that is
something which is captured by logics of
knowledge so again these are some very
interesting things which have come so
the paradigms that have come up from
knowledge and reduction and knowledge
based systems then there were a lot of
expert systems which had come up at some
point of time and we still have a lot of
systems which are able to use facts and
rules and
give you either expert opinion or
decision support systems then there are
automated zero theorem provers these
have there has been a phenomenal growth
in this area on automated theorem
provers and there are some really good
theorem provers which are there on the
net I will give you the links later on
and you can download and try out some of
them now in automated theorem proving
what happens is that you encode a
problem in logic and then you try to
solve it using the theorem prover and in
recent times there has been some
theorems which people could not solve
mechanically because of the size of the
proofs and we actually had automated
theorem provers doing the job for us and
notable among these is well it was not
exactly done with an automatic theorem
prover but one of the automated proofs
that we got in recent times was the
proof of the four-color theorem the
four-color theorem is you are given a
graph and you want to you are given a
planar graph and you want to see whether
you can color it with four colors and it
was an open question for a long time
that whether all planar graphs can be
colored with four colors or not and
people could not actually prove that
four colors are sufficient and then
people devised a way of generating the
kinds of configurations through which
you should be able to prove whether four
colors suffice or not and then this set
of configurations start growing so you
need an automated mechanism of
generating them and establishing the
proof
so the initial proof the
computer-generated proof was huge and
obviously mathematicians did not accept
it because they said that if we do not
have a way of checking whether the proof
is correct then what good is your
theorem proof so then actually we have
again a class of theorem provers which
takes a proof and tries to deduce
another proof which is humanly legible
so they try to talk together different
steps of the proof and try to give you a
proof with shorter number of steps so
there are theorem provers which attempt
to do that that kind of thing and
actually even for the four-color theorem
it was brought down to a shorter proof
which was then verified by humans
another area in which knowledge and
deduction is being used to a large
extent is formal verification this is an
area in which I work this is has found a
lot of application in verification of
circuits where you have huge circuits of
millions of gates and you have to be
able to verify whether the circuit is
going to do what you expect it to do and
there are techniques in formal
verification which will help you in
doing this so the main bottleneck that
we had in knowledge and deduction was
that the knowledge base or the fact base
can be potentially very large if you
look at the earliest chess-playing
computers they did not have much of a
knowledge base they just did a kind of
brute-force search we will study that
those kinds of searches that
game-playing programs do but later on
the more advanced chess-playing
computers rely on huge databases huge
knowledge bases they try to use as much
of prior knowledge as possible and this
has become possible today to a large
extent because storage has become free
so in the in the 90s to the 2000 storage
became free so we had huge discs
expensive storage mediums now you have
huge discs in the size of a matchbox and
they are also pretty fast so what is
going to happen in the next decade the
in the next decade communication will
become free now our main bottleneck is
communication here the bottleneck in on
your desktop you have
lots of space you have lost of
computation but communication is
becoming important because everything
you do not have locally you have the web
you have other kinds of resources all
around the world and communication is
going to become free in the next decade
so that next time when your washing
machine goes out of order you do not
have to bother it is directly going to
communicate with the dealer and put the
complain right so those kinds of things
are going to come and that is also going
to help the AI paradigm to a large
extent the third thing that we want to
look at is the ability to learn this is
another important aspect of artificial
intelligence so there we try to find out
that can we learn to solve a problem
better so we know how to solve a problem
we have device search algorithms we have
divided deductive algorithm for solving
a problem and then can we slowly start
learning the rules of the game and can
we learn to plan right so we are going
to study a little bit of learning in
this course specifically we will study a
little bit of what is called belief
networks and then we will also study
some parts of perceptrons and neural
networks and how we learn with these
kinds of technology so then what is I so
what we have seen so far is that we have
now at the current point of time a good
amount of power in terms of automated
problem solving a good amount of
understanding of machine learning and
also a fair amount of technology in
logic and deduction so this is the basic
computational background which is
necessary for AI to become real to make
a computer act as if it has intelligence
but that is not all what is coming up
and in
a very large way are these things
computer vision so we can equip the
computer to actually get a lot of
sensory inputs from the environment so
that we can get rid of keyboards and
Mouse's see these are very old-fashioned
things we are used to keyboards and
Mouse's but then there is no reason why
we should use something like a keyboard
if we can aim if we are able to talk to
the computer and make it read text from
whatever we write then things will
change then the ability to interpret not
natural languages so NLP this is another
area which is going to make a lots of
advancements in the forthcoming years
and then we have robotics where we
actually enable the computer to perform
realistic actions not just computational
actions but actions with objects and
with actual so what is actually
remaining in as part of this course is
human-computer interaction but since
this course is predominantly a computer
science course so we are going to focus
more on this side we are going to focus
more on automatic problem solving on
machine learning logic and deduction and
we are not going to really go into the
details of these but there are other
courses in which you can pick up inputs
about computer vision natural language
processing and robotics in this course
we will be sticking to this this side
so the fundamental things that we need
to have a background on is algorithms
but specifically what we are going to
learn from this course is how we can
actually express different kinds of
problems into our known algorithmic
frameworks and we will learn new
algorithms as well are you familiar with
gödel's incompleteness theorem so this
is the first thing that you want you
should read up okay so what this says is
in any language expressive enough to
describe the properties of natural
numbers there are two statements that
are undecidable that is the truth cannot
be established by any algorithm so you
are familiar with undecidable problems
have you done your course on flat formal
languages and automata theory so Turing
machines and stuff like that you know
like ulting problem of Turing machine
right so we are going to look at some
other undecidable problems like we will
see proving first-order logic formulas
etcetera where you have undecidability
creeping up again and we will see that
how this theorem gödel's incompleteness
theorem is actually an encompassing
theorem which says that if you are able
to express the properties of natural
numbers then this logic which is able to
do this will contain statements that are
undecidable right so you cannot have
approval which is going which will be
able to be everything including
reasoning about natural numbers right so
there will be some unsolvable things but
the ball does not stop there so we have
to still solve those problems so we will
see that how we can get around some of
these problems and still be able to do
things from an engineering point of view
then we have the church-turing thesis
which was proposed in 1936 which is that
the Turing machine is capable of
computing any com
durable function now this is the
accepted definition of computability and
there is no real reason why
computability is defined in this way but
that is what we have done so far that we
have actually defined computing in terms
of the compute ability of Turing
machines and that is accepted definition
with which we will go and then another
important thing which affected AI in a
large way was the notion of
intractability and NP completeness and
the and the concept of reduction now if
I have a problem P and I want to show
that this is np-complete I have a
problem say p1 and I want to show that
this is np-complete what do I have to do
so is it the case that I translate this
p1 to some known
is this what I will do reduce translate
p12 some known np-complete problem yes
or no how many say yes right it is
exactly the other way round what we need
to do is we need to take an np-complete
problem and then reduce that to an
instance of P 1 is that the case and why
because then you can use the following
argument that if I have a polynomial
time algorithm if I have a poly time
algorithm for solving P 1 then I can
solve this np-complete problem also in
polynomial time provided that I can do
this translation in polynomial time so
if I can translate this problem in
polynomial time to P 1 and I have the
polynomial time algorithm for solving P
1 then I have solved the empty complete
problem also in polynomial time so that
is why if we do if we are able to make
this translation then we can use this
argument and say that ok most likely we
will not have a polynomial time
algorithm for P 1 unless P is equal to
NP right
so the course outline is like this that
we start with problem solving by search
then under which we will look at state
space search problem reduction search
game playing etcetera then we will have
logic and deduction in which we will
study first-order logic
propositional logic a little bit of
temporal logic and deduction then we
will have a chapter on planning then we
will study a very interesting area of
reasoning under uncertainty where we
will have to deduce but then we are not
100% sure about the facts we are not
100% sure about the rules so we have to
perform the reasoning knowing that the
knowledge and the facts that we have
have some associated probability with
them then we will also study a chapter
on learning and a few additional topics
which during the course we will see
these are the references I will be
mostly following the book artificial
intelligence a modern approach by
Stewart Russell and Peter Novick there
is a Pearson LP low price Edition for
this and then I will also look up a
little bit of principles of artificial
intelligence of Nilsen now both of these
books are there in sufficient numbers in
your departmental library
so today's class is the first chapter
that we are going to take up in this
course this the topic is problem solving
by search such this slides so under
problems are window search we will look
at cell deform search firm works the
first is what we will call state space
such so here we will see how a real
world problem can be mapped into a state
space and then we can have such
algorithms for searching out the goals
in the state space so under this we are
going to look at three different types
of search namely uninformed search or
blind search these are search problems
where we do not know any domain specific
information so we have to search without
any additional knowledge and the second
thing will be informed of realistic
search where we will have functions
provide the search procedure and we may
also study some algorithms firing front
heuristic search the next topic that we
will study is public deduction search
this will be done in the next lecture
and this will talk about problems which
can be decomposed into subproblems in a
planar
to dynamic programming except then we
will try to understand through search
how best to solve the problem so take
for example integration by parts you
remember integration by part so we can
solve the same problem in different ways
so we will associate a cost function
with every kind of solution and then try
to see what's the best way to decompose
the problem into parts and solve it the
best okay so that is what we in one in
problem reduction search and finding
even have a look at game tree search to
give you an idea about how chess-playing
programs work so we will have a few
lectures on searching game trees and see
how you can be atomization in the
presence of an adversary and I will
briefly touch upon some advances in the
search area which is mostly research
done in our department by our AI group
and this is a memory bounded search
multi objective search and learning how
to search so these will be just briefly
touch upon what these topics are right
so coming to state space search the
basic search problem is as follows know
given a for purpose the foot couple the
filter by the first is s which is the
set of states which is implicitly
specified now this is very important if
you have a a graph and you are asked to
find out the shortest path from one node
to another node in a graph well we know
that you can do it with Dijkstra's or
any other shortest path algorithm and
the problem is not
we have polynomial time algorithms are
doing otherwise if you apply something
like depth-first search and in DFS if
you cannot determine whether you visited
some state then you can potentially go
into an infinite loop so we will have to
devise mechanisms by which we are able
to check whether we are visiting a sense
state all we are arriving in the same
stage along some other part it and so on
so now you look at their outline of our
first basic search algorithm and you
will develop this search algorithm to
encompass all the different kinds of all
search algorithm that we will study so
we will start by maintaining a list
called open this is a business
terminology F F a list called open was
the best many B caves back and so we
will continue the same names and zoggles
so initially I will be straighten will
contain the start state this is the
first initialization then we have one
termination criterion if we find that
the list open is empty then we will turn
benefit failure
let us look at a little bit more than to
become clear third step is we select a
node n from open we select one state n
from open and then you check whether n
is above if M is a goal then we have
found a bill shall we terminate with
success remember though that our
objective is to start from the start
step and find the path previously so if
n belongs to G then we terminate with
success otherwise we generate the
successors of n using the transition
operators oh and insert these new States
in open so what we are doing is we are
starting from one state and then
expanding that state to generate its
successor States by using the state
transition operators and inserting these
States into effect so at this point of
time I have not mentioned where open is
a FIFO ever before or whether it is a
parity ki netic and these algorithms
will vary depending on the kind of data
structure that we use person but this is
can only work if that state transition
operators are reversible so that you can
start from the build and work backwards
and in puzzles like 15,000 etcetera it
is indeed reversible so you can climb up
from website of the bi-directional
search where we can start
then if you did normal depth first
search then you will go straight from
this start step and expand all nodes up
to deck B so your complexity will be
order of B to the power of B here we are
doing is we are starting from the start
state and moving this direction and we
are also starting from the goal state
and moving in the other direction using
the reverse of the state transition
operators and we continue until the
frontiers of the searches from these two
sides meet at some point of time at some
place when you have a common node in
these two then you can find out a path
from the start state to the goal State
now we see that because let be one
mutation from this side one iteration
from that side so you will be e to the
power of T patients from this side and D
to the power of P patients from the
other side right order of D to the power
of Q from this side already to the power
two from the other side if you do B to
the power T tration from this side where
complexity for DFS is order of B to the
power D by two and also from the other
side it is order of B to the power of D
by 2 and if you add up this Plus this
we still have order of B to the power of
D by 2 right so you have actually made
their complexity reduce the complexity
by half Afra of B to the power of D of
order B power ultimately be significant
actually right so is complete this
lecture here and from tomorrow onwards
we will start looking at this set at
this end research problem with the
introduction of costs so we will see
what happens when the step transition of
letters have associated cost when the
bills are associated cost
19:51 a young nation aspiring to find
ways to fulfill urdu Numa lays the
foundation of an institution that will
give aspiring technic rats the license
to fly high the first Indian Institute
of Technology is born at correct board
founded on the basis of the
recommendations of the NRS Sankar
committee that was set up in 1945 to
consider the development of higher
technical institutions in India the
Institute was first established in five
esplanade East Kolkata before the move
to curb born in 1951 with Sir Garin
Chandra Ghosh as the first director and
dr. BC Roy as one of its founding
Guardians the Institute established
itself as the symbol of a young dynamic
and resurgent nation as top students rub
shoulders with the most celebrated of
professors and scholars
visions took shape and IIT kérékou
continued to play the pioneering role
that was envisaged for it
enabling India to become a knowledge
powerhouse that it is today
at every stage of its evolution iit
carrot will remained ahead of its times
it provided the best of facilities for
the budding technologists helping them
shape their own as well as the nation's
future indeed today iit carrot pool has
blossomed into a time-tested venerable
Institute of learning with the rich
experience of converting individuals
into brilliant professionals through 50
glorious years as you cross the campus
gate you feel the distinct NIP that is
iit kharagpur
the spirit of objective inquiry and
lateral thinking hangs heavy in the air
the modern township like camper society
color pools set in Sylvan surroundings
is self-sufficient in all respects from
modern bangs to the good old post office
from vast playgrounds in well-equipped
jeans to modern auditoría and open-air
theaters
from the quiet fiber-optic loot
residential quarters for the faculty to
the web-enabled hostel rooms for the
students at iit carpool lush green
bowers of tranquility coexist with smart
cards and ATMs spread over six hundred
and ninety hectares of sprawling cyber
habitat a hundred and twenty kilometers
from kolkata iit kharagpur
is one of the largest network campuses
in asia just the academic complexes
itself spreads over a built-up area of
about two million square feet of which a
hundred and fifty thousand square feet
is the new complex that commemorates the
golden jubilee celebrations and that's
not all it is the only IIT to have
conquered territory beyond its own
through cutting edge courses offered in
its extension campuses at Kolkata and
Ruben issuer IIT Kharagpur is not just
about its large campus but its diverse
range of activities it offers the widest
spectrum of disciplines ranging from
aerospace
biotechnology cryogenics - architecture
mining and agricultural engineering
supported by strong faculties of
sciences humanities and management there
are more than 30 departments and centers
that offer the largest number of
undergraduate and postgraduate courses
amongst the IITs the courses are
ever-evolving and show the way for other
sister institutions the richness in its
diverse activities is showcased by the
technological support the institution
provides in areas like architecture
agriculture post-harvest technology and
medical sciences
the institute has revolutionized and
popularized rice milling technology the
other major contributions of iit
kharagpur have been in the critical
fields of Defence railways space
research Power Systems and
petrochemicals all these activities
directly empower the human requirements
of the nation advanced facilities at the
Institute make it possible to undertake
cutting-edge research and service
sponsored research projects the array of
equipment ranges from aerodynamic
testing laboratories to intelligent
machining centers atomic spectrometers
to VLSI design labs molecular beam
epitaxy to anechoic chamber farce
protein liquid chromatographs to liquid
nitrogen plans
the cutting-edge technologies are at par
with the best research facilities across
the globe in fact the volume of research
and development activities at the
Institute is incredible in terms of the
number of patents it owns the volume of
industrial consultancy it provides and
the revenue that it hangs from all these
make iit kharagpur a class apart and
strengthens its position as the true
pioneer in technological education in
India the Institute library deserves a
special mention fully web-enabled it is
one of the largest in Asia with over
three hundred and twenty four thousand
volumes of material including ebooks
videos microphone and baton documents
Bin's that ensure a student's mind
develops at the right pace along with
its strong sense of academics which is
ensured by a strict selection process
life at Carrick pool is a celebration of
well life and at its heart are the
students in fact the same goes that you
can take an IIT and out of k GP but not
gay GP out of an IIT n you've left a
part of you back
for most of the students life in the
campus was in itself a festivity a
collage of activities that shape their
mind and body a collage of events that
was a synthesis of competition and
cooperation a collage of interests as
diverse as dramatics and AM radio yes
life at Caribou has always been exciting
and the years cemented lifelong bonds as
lives mingled over cups of joy and
stretched a row stimulating semesters
the holes with their blocks and wings
connected by charming catwalks remain
ensconced in their own world a collage
of memories infrastructurally adequate
architectural a meticulous and
holistically inspiring where students
wherever they might be from invariably
come to their own developing their
individual talents honing their skills
to take on challenges with confidence so
they can move ahead in fulfilling their
dreams
what makes iit carrot bulls so unique is
its environment undiluted by the
diversions of metropolitan surroundings
the close-knit campus life enhances the
entrepreneurial and innovative spirit of
the achievers to be in an environment
that is so stimulating it is only
natural that down the years iit
kharagpur has consistently produced
well-rounded individuals many of them
are celebrities
their own right holistic grooming has
had a lot to do with this so no matter
which walk of life they choose the IIT
kgp ID stands tall and so does the
institution that bred him in majestic
splendor the alumni of this institute
command global respect their
distinguished presence at the helm of
global joins is a matter of national
pride for the students of IIT Kharagpur
it is impossible to erase any scratch of
memory about their alma mater in fact
some come back to invest sentiment pride
and money to see the Institute they call
home rise to even greater heights
structurally functionally as well as
holistically their singular aim is to
make iit kharagpur and Institute
whichever way you look at it our
excellence
a man's journey into quiet
accomplishment and the hall-of-fame
starts with the right step and the
training and placement cell of IIT has
will be efficient facilitator in this
regard for over 30,000 graduates having
placed more than 95% of its students in
a wide range of industries consistently
for over two decades it is no wonder
that the Institute is the most preferred
campus for technical recruitment of
quality manpower with infrastructure
like industrial power and communication
facilities in addition to its excellent
research in consultancy facilities step
or science and technology entrepreneurs
Park aims to assist the budding
entrepreneur into a successful adventure
capitalist guiding him right from the
concept institutional financing
production leading up to the launch and
marketing of the product with its rich
pool of talent and excellent
infrastructure it is no surprise that
through the last three decades iit
kharagpur
has developed strong layers of would-be
industry
sr9c or sponsored research and
industrial consultancy sell was
formalized as the special industry
interaction cell in 1982 devoted full
time to handle industrial projects and
consultancies and for deploying and
propagating intellectual property
successful sponsored research projects
straddle a wide spectrum ranging from
computers communication and
biotechnology to robotics photonics and
food processing
the setting up of a state-of-the-art
VLSI CAD laboratory and tie-ups with GE
in areas ranging from vehicle structure
design to electrical communication and
software technologies are excellent
examples of IIT characters ever evolving
pioneering spirit collaborations with a
host of national and industrial majors
are a testimony of its proven expertise
and research repertoire
Oh
indeed in this Golden Jubilee year as
the celebration continues pundit Nehru
would surely have been a proud man today
for him
IIT Kharagpur was always more than just
an Institute of Technology in his own
immortal words it is indeed a fine
monument of modern India
you
you
you