Application Form Application No:

2022PHDDECMA0187
December 2022
IIT Guwahati

Application Details
Academic Programme: PhD
Department: Mathematics
Project Code (If Any):
Disciplines/Specialization: 1. Probability, Statistics and Finance
2. Mathematics
3.
Research Sub-Areas (If Any): 1. Information Theory
2. Linear algebra and Functional Analysis
3.
Project Mode Programme (If Any):
JDP with IIT(BHU) (If Any):

Personal Information
Candidate's Name: GAURAV KANDPAL
Father's Name: BHUWAN CHANDRA KANDPAL
Gender MALE Date Of Birth 28/01/1998
GENERAL
Marital Status SINGLE Caste Category
EWS
Nationality INDIAN Studentship Type Regular
Physically Disabled (PwD) No Nature of disability
Percentage of your disability (%) IFSC code of the Bank SBIN0017342
Savings Bank Account Number of the Nearest Railway Station to the place of your Correspondence
34855040652 HALDWANI
Candidate: Address

Correspondance Address Details Permanent Address Details

GOLNAKARARIYA
GOLNAKARARIYA
NEAR ANAND HOTEL
NEAR ANAND HOTEL
ALMORA
ALMORA
ALMORA
ALMORA
UTTRAKHAND, PIN: 263601
UTTRAKHAND, PIN: 263601
Email: [email protected]
Mobile: 9528373148
Mobile: 9528373148

Qualification Details
Details of Class X or Equiv. and Class XII or Equiv.#
Total Marks Year of
Examination University/College/Board Percentage Subjects
Marks Obtained Passing
UTTRAKHAND STATE MATHEMATICS, SCIENCE,
X 500 457 91.4 2013
BOARD ENGLISH
UTTRAKHAND STATE MATHEMATICS, PHYSICS,
XII 500 455 91 2015
BOARD CHEMISTRY

Degree Details

Name of the Degree B.SC. HONS MATHEMATICS Name of the College/Institute DELHI UNIVERSITY
Degree awarded by College/Institute BACHELOR IN SCIENCE Discipline/Major/Branch MATHEMATICS
Specialization(if any) Status PASSED
Year of Joining 2015 Month and Year of Passing June-2018
Grading Format CGPA/CPI Percentage of Marks/CPI Obtained 8.12
100%/Max. CPI 10 Marks grades available for uploading YES
Have you done any courses on Mathematics on Bachelor degree level? YES
Do you have consistent First Class in all academic qualifications? YES
Master Degree Details

INDIAN INSTITUTE OF
Name of the Master Degree MSC Name of the College/Institute
TECHNOLOGY PALAKKAD
Degree awarded by
MASTER OF SCIENCE Discipline/Major/Branch MATHEMATICS
College/Institute
INFORMATION THEORY AND
Specialization(if any) Status PASSED
STATISTICS
Year of Joining 2020 Month and Year of Passing June-2022
Percentage of Marks/CPI
Grading Format CGPA/CPI 8.13
Obtained
Marks grades available for
100%/Max. CPI 10 YES
uploading

GATE/NET/CEED Details

Is Your Qualifying Degree from IIT? Yes Name of IIT IIT Palakkad CPI/CGPA of IIT 8.13
Qualified in National Level Scholarship Exam? No Name of the National Level Scholarship Exam Qualified No

Referee Details

Referee Name Designation Affiliation/ Institute Email Phone

M. ASHOK ASSISTANT INDIAN INSTITUTE OF TECHNOLOGY
[email protected] 7223062058
KUMAR PROFESSOR PALAKKAD

Uploaded Documents

Sl No. Document Details Download

1. Bachelor Degree Certificate and Marksheet Uploaded
2. Caste Certificate Uploaded
3. Class 10 Certificate Uploaded
4. Class 10 Marksheet Uploaded
5. Class 12 Certificate Uploaded
6. Class 12 Marksheet Uploaded
7. Date Of Birth Certificate Uploaded
8. Master Degree Certificate and Marksheet Uploaded
9. One page Write up on Proposed Research Work Uploaded

10. Photo

11. Sign

Fee Payment

Transaction No. Amount Payment Status

2022PHDDECMA0187 300.00 Success!
Scanned with CamScanner
Scanned with CamScanner
Scanned with CamScanner
Scanned with CamScanner
Scanned with CamScanner
Scanned with CamScanner
Scanned with CamScanner
 No. 000633
No.212005016 Sr.
3tgaoaicr/Roll

INDIAN INSTITUTE
OF TECHNOLOGY
PALAKKAD

USYTT

f faarcs

The Board of Governors

of the
IndianInstitute of Technology Palakkad
upon the recommendation of the Senate
hereby confers on

Gaurav Kandpal1
the degree of
Master ofScience
in

Mathematics
Given on this day, the Thirtieth of July, Two Thousand andTwenty Two
under the seal of the Institute

Rogistrar Chalrman, Senate and Director Chaiman, Board of Govemors

 Indian Institute ofTechnology Palakkad
Nurturing Minds For a Better World IIT PALAKKAD
CUMULATIVE GRADE CARD
Name :GauravKandpal Roll No :2120050016

Program: M.Sc. (Mathematics) USID :244b5462-e87b-42fc-b22a-69acbea96077

Code Course Title Cat Cr Gr Att Code Course Title Cat Cr Gr Att

SemesterI Semester I

MA5001 Linear Algebra PST VG 1 MA6005 Functional Analysis PST 4 B VG

MAS003 Real Analysis PST VG MA6007 Partial Differential equations PST

PST C VG 3
)
3 MA5005 PSE-
Groups and Rings CS5512 Machine Leaming

4 MAS007 Probability and Statistics PST 4 A VG 4 MAS110 Project (Level PSP 3 B VG

MA5009 PST 4 B VG 5 PH5619 of OE C VG
5 Ordinary Differential Theory and Applications
Stochastic Processes
Equations
6 GNS000 Communication and CWC 0 Y VG
Technical Writing Skills

Semester II Semester IV
PST B VG MA5120 PSP VG
MA502
2 MA5004
Multivariable Calculus

Measure Theory PST

4

4 B G
1

2 css014
Project (Level I)

Foundations of Data Science PSE 3A VG

and Machine Learning

3 MA5006 Fields and Modules PST 4 B G 3 EE5517 Information Theory and PSE 3 A G
MA5008 Topology
PST 4 C G4
VG
GN6001
Statistics

Research Methodology
Professional Ethics
and CWC 0 Y VG

5 MAS010 Complex Analysis PST

Unique Student Identification Number

USID

Semester
20 20
Total credits
20 20
Earned credits
8.00 8.00 9.0
GPA 7.8
.95 8.13
CGPA 8.00 8.00

Place &Dateof Issue: Palakkad, 15-07-2022

AN
INS
OTE
OFS

ADO1
officer

Indian
In-Charge(Academics)

Dr.B.Thlagarajan/st.
Deputy Registrar
Institute of Technology
Ahalia integrated Campus, Kozhipara
04923 226 521
. fdurTTT

Palakkad

Palakkad 078857.Ph:
 Guessing, Source coding and Tasks partitioning in
distortion case
1 STATE OF THE ART AND PROPOSED PLAN
Consider the following game: Bob draws a sample X (a random variable) from the alphabet set X with PMF p. Then,
Alice who does not see X, will try to guess it by making a strategy. The goal is to minimize average number of attempts
required to guess it. Let G(x) denote the number of attempts to guess x correctly by a particular guessing strategy. In
1994, Massey [2] observed that E[G(X)], the average number of successive guesses, is minimized by a guessing strategy
that guesses the possible values of X in decreasing order of probabilities p. For an infinite set X , he gave a lower bound
in terms of Shannon entropy H(p). He also showed that there is no interesting upper bound in terms of H(p). In
1998, Arikan [3] studied the guessing problem subject to distortion. He showed that, if we allow some distortion with
a distortion level D > 0, then also we can bound the ρth moment of guessing function in terms of guessing exponent
E(D, ρ), which is defined in terms of rate distortion function and relative entropy.
In 1948, Shannon [1] introduced source coding problem. In this problem, he tried to encode an alphabet set X in an
uniquely decodable manner. By a uniquely decodable code, we mean a code whose lengths satisfy the Kraft inequality.
The goal of Shannon was to find bounds for expected value of lengths L(x) of the code for x ∈ X , E[L(X)]. He found
a lower bound for E[L(X)] in terms of Shannon entropy H(p). In 1965, Campbell [12] studied a more general coding
problem for finite alphabet set. Campbell found lower and upper bounds for normalised ρth order cumulant in terms
of Rényi entropy Hα (p) of order α = 1/(1 + ρ). Shannon also introduced source coding problem with distortion (Lossy
source coding) in his 1948 paper. Arikan [3] studied lossy source coding in 1998 and found some relationship between
guessing with distortion problem and lossy source coding problem. Further in 2010, Sunderasan [7] showed this relation
in an asymptotic sense in a technical report.
In 2014, Bunte and Lapidoth [5] studied a new problem called “Tasks Partitioning Problem”. In this problem, we
have a finite set of tasks X . Suppose a task X is randomly chosen from X according to a probability distribution p.
These tasks are associated with M number of keys. The problem arises when we have limited number of keys, that is,
number of tasks is greater than number of keys. So we have to assign more than one tasks to some keys to perform all
tasks. When we want to perform a task, the key associated with it is pressed. Consequently, all the tasks associated
with this key will be performed. The Goal of the problem is to assign tasks to each key in such a manner so that the
number of redundant tasks will be minimum. This problem is same as partitioning the set X in M partitions, such that
the expected value of partition function E[A(X)] will be minimum, where A(x) gives the cardinality of the subset from
partition, where x belongs. If there is no distortion allowed, Bunte and Lapidoth [5] showed that there exists a sequence
of partition functions such that the ρth moment of the sequence of partition functions converges to 1. If there is some
allowed distortion D > 0, then while performing a task x, if we press the key associated with x, the key will perform
tasks x̂ satisfying d(x, x̂) ≤ D and some redundant tasks associated with it. In distortion case, we get the partition of
the set of obtained tasks Xˆn . Bunte and Lapidoth showed that there exists a sequence of partition functions with size at
most Mn (a sequence of positive integers), such that the ρth moment of the sequence of partition functions converges to
1, whenever limn→∞ Mn /n =: γ exists and is greater than E(D, ρ)/ρ. Further, he proved that, if γ < E(D, ρ)/ρ, then
the normalised ρth moment of a sequence of partition functions with size at most Mn converges to E(D, ρ) − γρ. Here
E(D, ρ) is the same guessing exponent, which we have in the case of guessing with distortion.
All three problems show some similarities in distortion case in an asymptotic nature of solution. In [3], [5] and [7], there
are some results to prove a unified relationship among all three problems in distortion case. The research over finding
such relationship, can help to find solution of one harder problem of Tasks partitioning or Guessing with distortion by
finding a solution of corresponding Source coding problem in distortion case. Moreover, it is also interesting to work on
these problem with Large deviation approach which may give new results in this area. In 2008, Hanawal [13] worked
on Guessing problem without distortion considering Large deviation approach. Similar results can be proved for Source
coding problem and Tasks partitioning problem in distortion case.

References
[1] Shannon, C.E. (1948), A Mathematical Theory of Communication. Bell System Technical Journal, 27: 379-423.
[2] J. L. Massey, “Guessing and entropy,” in Information Theory, 1994. Proceedings., 1994 IEEE International Symposium
on. IEEE, 1994, p. 204.

[3] E. Arikan and N. Merhav, ”Guessing subject to distortion,” IEEE Transactions on Information Theory, vol. 44, pp,
1041-1056, May 1998.
[4] E. Arikan, ”An inequality on guessing and its application to sequential decoding,” IEEE Transactions on Information
Theory, vol. 42, no. 1, pp, 99-105, Jan 1996.

1
 [5] C. Bunte and A. Lapidoth, “Encoding tasks and Renyi entropy,” ´ IEEE Transactions on Information Theory, vol.
60, no. 9, pp. 5065–5076, Sep. 2014.

[6] R. Sunderasan, ”Guessing based on Length Function,” IEEE Transactions on Information Theory,
[7] M.K.Hanawal and R.Sundaresan, Guessing and compression subject to distortion, Technical Report TR-PME-2010-
12, 23 February 2010.
[8] T. M. Cover and J. A. Thomas, Elements of Information Theory, 2nd ed., Wiley, 2012.

[9] I. Csiszar and P. Shields, ´ Information Theory and Statistics: A Tutorial, Foundations and Trends in Communications
and Information Theory. Now Publishers, 2004.
[10] R. Gallager, Information Theory and Reliable Communication. New York, NY, USA: Wiley, 1968.
[11] I. Csiszar and J. Kórner, Information Theory: Coding Theorems for Discrete Memoryless Systems. New York:
Academic, 1981.

[12] L. L. Campbell, “A coding theorem and Renyi’s entropy,” ´ Information and Control, vol. 8, no. 4, pp. 423–429,
1965.
[13] M. K. Hanawal and R. Sundaresan, Guessing Revisited: A Large Deviations Approach 2008, DRDO-IISc Programme
in Mathematical Engineering Technical Report No. TR-PME-2008-08.

[14] Kumar, M. Ashok and Sunny, Albert and Thakre, Ashish and Kumar, Ashisha and Manohar, G. Dinesh, “Are
Guessing, Source Coding, and Tasks Partitioning Birds of a Feather?”, 2020

Powered by TCPDF (www.tcpdf.org)

2