.01

ABOUT

PERSONAL DETAILS

BIO

ABOUT ME

My recent research interests have revolved around networking and machine learning. Recent works include incremental deployment of class based hybridization models for campus networks and real time monitoring of QoS metrics in Software Defined Networks (SDN). I am also working on lung cancer detection and ear biometrics leveraging CNNs and ResNets.

My work on leveraging WebRTC for P2P content distribution in web browsers and character recognition in natural scene images have been accepted at international conferences. I have been among top 25% at various data science challenges (Kaggle.com) and completed 12 MOOCs on data science and related.

I am a national level percussionist in Tabla, drums, octopad etc.

I want to continue working in the area of SDN and I am looking for collaborative research and development opportunities.

NEWS

TALKS & PRESENTATIONS

  • Apr 3, 2017: Presenting a measuring control packets in SDN at SOSR'17, CA, US.
  • Sept 22, 2016: Presenting a WebRTC + Chord Framework at ICACNI'16, NIT, Rourkela, IN.
  • Mar 23, 2016: Presenting a new P2P Framework using WebRTC at AINA'16, Switzerland.
  • Dec 23, 2014: Presenting at Winter School on Character Recognition at IIITD, N.Delhi.
  • Aug 01, 2012: KVPY Fellowship awarded by DST, Govt. of India.

HOBBIES

LEISURE

I love to play tabla and listen to Indian classical music in my leisure.
Participated as vocalist, dancer and percussionist at National Level Youth Festivals. I have played Tabla, karlaals, tasha, drums, octopads, triple etc. I have performed with following dance forms: koli, rajsthani, chhau, nagpuri, patriotic etc.

INTERESTS

RESEARCH

SDN
P2P Networks
Data Science/Computer Vision
Comp. Architecture

.02

RESUME

  • EDUCATION
  • 2016
    2018
    Pilani

    M.E. Computer Science

    BIRLA INSTITUTE OF TECHNOLOGY AND SCIENCE

    Masters degree at the Department of Computer Science and Information Systems, BITS Pilani
  • 2012
    2016
    Pilani

    M.SC. (TECH.) INFORMATION SYSTEMS

    BIRLA INSTITUTE OF TECHNOLOGY AND SCIENCE, 8.1/10

    Undergraduate degree at the Department of Computer Science and Information Systems, BITS Pilani
  • 2004
    2011
    Hazaribag

    SANGEET PRABHAKAR

    PRAYAG SANGEET SAMITI, ALLAHABAD.

    B.A. (equivalent) degree with First Div. and Distinction for recognition in instrumental percussion
  • ACADEMIC AND PROFESSIONAL POSITIONS
  • 2013
    2014
    Pilani

    Graduate Teaching Assistant

    CLICKCOUNSELOR

    • I assist the ARC division in automation of maintenance of students' academic history, the usage of various academic flexibilities available to the students.such as registration, addition, substitution, withdrawal etc.
    Pilani

    CO-FOUNDER-CUM-TECH. LEAD

    CLICKCOUNSELOR

    • Developed a browser based video conferencing software using WebRTC
    • An effort to bring the ecosystem of Counselees and Counsellors online and conduct online counselling via video conferencing
    • Selected for implementation at Pilani Campus, BITS Pilani. Finalist at Vishisth, All India B-Plan competitiion, IIT Delhi & Ventura, NIT Trichy
  • 2014
    Ahmedabad

    SOFTWARE DEVELOPMENT INTERN

    ELITECORE TECHNOLOGIES

    • Hadoop Deployment & Study of Hadoop packages in Bioinformatics: Hydra, Chorus and SeqPig with comparison of Hadoop and ApacheSpark.
    • Developed a WP App to locate and connect to nearby Wi-Fi Hotspots
  • HONORS AND AWARDS
  • 2009
    2010
    Bangalore

    Merit Scholarship cum. 40% Fee Wavier

    OFFERED ONLY TO 5% OF THE HIGHER DEGREE STUDENTS

    awarded by BITS, Pilani
  • 2009
    2010
    Bangalore

    KVPY SCHOLARSHIP

    NATIONAL AWARD FOR PURSUING RESEARCH

    awarded to top 125 students from the country for research by Department of Science and Technology (DST), Govt. of India
  • 2011
    NEW DELHI

    CBSE NATIONAL LEVEL SCIENCE EXHIBITION

    NATIONAL AWARD FOR SCIENCE PROJECT

    One of few projects selected from CBSE Patna Zone for representation at national level. Adjudged as the best presentation in state.
.03

PUBLICATIONS

PUBLICATIONS LIST
ALL
SORTING BY DATE
08 MAY 2017

MPLS BASED HYBRIDISATION IN SDN

Proceedings of 4th IEEE International Symposium on Software Defined Systems IEEE-SDS 2017, Valencia, Spain

We use the standard MPLS data-plane together with a control-plane based on OpenFlow to come up with a Hybrid SDN model
Conferences Accepted Yash Sinha, Siddharth Bhatia, G S S Chalapati, Virendra Shekhawat
img

MPLS BASED HYBRIDISATION IN SDN

Yash Sinha, Siddharth Bhatia, G S S Chalapati, Virendra Shekhawat Conferences Accepted

We use the standard MPLS data-plane together with a control-plane based on OpenFlow to come up with a Hybrid SDN model

03 APR 2017

METICULOUS MEASUREMENT OF CONTROL PACKETS IN SDN

Proceedings of the ACM Symposium on SDN Research, SOSR '17, pp. 187-188, Santa Clara, CA

The data packet statistics sent by OpenFlow compliant switches cumulatively includes statistics about control traffic which is used for network control and management. This reduces the accuracy of calculation of QoS metrics and thus hampers network monitoring. We present here a novel algorithm to accurately measure the fraction of control packets in SDN within 3% error rate.
Poster Accepted Yash Sinha, Shikhar Vashisth, K. Haribabu
img

METICULOUS MEASUREMENT OF CONTROL PACKETS IN SDN

Yash Sinha, Shikhar Vashisth, K. Haribabu Poster Accepted

The data packet statistics sent by OpenFlow compliant switches cumulatively includes statistics about control traffic which is used for network control and management. This reduces the accuracy of calculation of QoS metrics and thus hampers network monitoring. We present here a novel algorithm to accurately measure the fraction of control packets in SDN within 3% error rate.

22 SEP 2016

A BROWSER-BASED DISTRIBUTED FRAMEWORK FOR CONTENT SHARING

Progress in Intelligent Computing Techniques: Theory, Practice, and Applications. Vol. 1 & 2. Ed. by Pankaj Kumar Sa, Manmath Narayan Sahoo, M. Murugappan, Yulei Wu & Banshidhar Majhi. Vol. 518 & 519 of Advances in Intelligent Systems and Computing. Springer Singapore, 2017.
4th International Conference on Advanced Computing, Networking, and Informatics (ICACNI), NIT, Rourkela, India, 22 nd - 24 th Sept 2016.

We propose a browser based, asynchronous framework of a P2P network using distributed, lookup protocol (Chord), NodeJS and RTCDataChannel; which is scalable and lightweight. The design combines the advantages of P2P networks for better and sophisticated education delivery. The framework will facilitate students to share course content and discuss with fellow students without requiring any centralized infrastructure support.
Conferences Accepted Shikhar Vashishth, Yash Sinha, K. Haribabu
img

A BROWSER-BASED DISTRIBUTED FRAMEWORK FOR CONTENT SHARING

Shikhar Vashishth, Yash Sinha, K. Haribabu Conferences Accepted

We propose a browser based, asynchronous framework of a P2P network using distributed, lookup protocol (Chord), NodeJS and RTCDataChannel; which is scalable and lightweight. The design combines the advantages of P2P networks for better and sophisticated education delivery. The framework will facilitate students to share course content and discuss with fellow students without requiring any centralized infrastructure support.

23 MAR 2016

ADDRESSING CHALLENGES IN BROWSER BASED P2P CONTENT SHARING FRAMEWORK USING WEBRTC

IEEE 30th International Conference on Advanced Information Networking and Applications (IEEE-AINA) pp. 850 – 857., Le Regent Congress Centre, Crans-Montana, Switzerland (CH), 23rd - 25th Mar 2016. Co-organised by HES-SO University & Icare Research Institute.

In this work, we investigated the feasibility of a browser based P2P content sharing framework based on Chord and WebRTC. We tried to couple the benefits of peer to peer (P2P) architecture (elimination of centralized dependency,better scalability, shareability etc.) along with the advantages of recent web technologies (NodeJs, WebRTC etc.) together.
Conferences Acceptance Rate: 29% Shikhar Vashishth, Yash Sinha, K. Haribabu
img

ADDRESSING CHALLENGES IN BROWSER BASED P2P CONTENT SHARING FRAMEWORK USING WEBRTC

Shikhar Vashishth, Yash Sinha, K. Haribabu Conferences Acceptance Rate: 29%

Certain issues such as lack of full fledged threading/concurrency support in the Javascript language, reliance on synchronous loading etc. restrict modern day browsers to take full advantage of current day multiprocessing capabilities. In this work, we investigated the feasibility of a browser based P2P content sharing framework based on Chord and WebRTC. We tried to couple the benefits of peer to peer (P2P) architecture (elimination of centralized dependency,better scalability, shareability etc.) along with the advantages of recent web technologies (NodeJs, WebRTC etc.) together.

In particular our contribution was the following:
● proposed browser based structured p2p network, delineating use of WebRTC
● proposed certain changes required in the Chord protocol to make it asynchronous and compatible with WebRTC
● specific features for LAN/WAN networks to speed up resource lookup and WebRTC offer exchanges
● prototypical implementation on LAN and extensive tests for functionality and scalability
● evaluation on the Internet by connecting a total of 30 peers on eight Amazon EC2 instances located at eight locations across the globe.

2 OCT 2015

COMPARATIVE STUDY OF PREPROCESSING AND CLASSIFICATION METHODS IN CHARACTER RECOGNITION OF NATURAL SCENE IMAGES

Machine Intelligence and Signal Processing. Ed. by Richa Singh, Mayank Vatsa, Angshul Majumdar, & Ajay Kumar. Vol. 390. Advances in Intelligent Systems and Computing. Springer India, 2016. pp. 119 – 129.
IEEE SPS-ASPIPA International Workshop on Machine Intelligence and Signal Processing (MISP), IIITD, New Delhi, India, 20 th - 23 rd Dec 2014.

A pipeline of Image Processing and Machine Learning algorithms to recognize characters segmented out from the Natural Scene Images.The challenge in developing solution for this problem was due to the occlusion,color and noises in the Natural Images. So a proper Image processing pipeline is needed to overcome such challenges and provide feature data for Classifier in order to classify correctly. The pipeline used included binarization of the image using Otzu thresholding,noise filters,Morphological filters. Used HoG to extract features and extra tree classifier for classification. The solution was able to achieve an accuracy of around 73%.
Book Chapter Yash Sinha, Prateek Jain, Nirant Kasliwal
img

COMPARATIVE STUDY OF PREPROCESSING AND CLASSIFICATION METHODS IN CHARACTER RECOGNITION OF NATURAL SCENE IMAGES

Yash Sinha, Prateek Jain, Nirant Kasliwal Book Chapter

Essentially, we proposed a classification technique for classifying characters based on a pipeline of image processing operations and ensemble machine learning techniques. The framework comprised of a sequence of operations such as resizing, gray scaling, thresholding, morphological opening and median filtering on the images to handle background clutter, noise, multisized and multioriented characters and variance in illumination. Image pixels and HOG (Histogram of Oriented Gradients) were used as features to train three different models based on Nearest Neighbour, Random Forest and Extra Tree classifiers to break the benchmark set by Fraz et al.[2], from 72% recognition accuracy to 73.17%.

[1] deCampos, T.: The Chars74k Dataset 2009. 15 Feb 2016.
[2] Fraz, et al. "Exploiting colour information for better scene text detection and recognition." International Journal on Document Analysis and Recognition (IJDAR) 18.2 (2015): 153167.

<
.04

RESEARCH

RESEARCH PROJECTS

REALIZING HYBRIDIZATION IN SOFTWARE DEFINED NETWORKS

INCREMENTAL DEPLOYMENT OF CLASS BASED HYBRID SOFTWARE DEFINED NETWORK

The migration to hybrid SDN, especially for campus networks causes both deployment and operational issues. Therefore, a systematic incremental deployment methodology as well as a hybrid operation model is needed. This work attempts to realize implementation and assess performance of a class based hybrid SDN model. We propose a design model for class based hybridization in SDN and implementation of a prototype.

ESTIMATION OF RAW PACKETS IN SDN

ESTIMATION OF THE FRACTION OF CONTROL, NON-USER GENERATED TRAFFIC IN SDN

A method to estimate the fraction of the network traffic that consists of raw packets in Software Defined Networks. The number of raw packets transferred depends on the number of switches & hosts in the network and it is a periodic function of time. Through experiments on several network topologies, we have estimated a way to find a cap on the generated raw packets in the network, using spanning tree information about the topology.

REAL TIME ESTIMATION OF PACKET LOSS IN LARGE SCALE NETWORKS USING SOFTWARE DEFINED NETWORKING (SDN)

PER PORT & PER FLOW PACKET LOSS ESTIMATION BY POLLING OPENFLOW COMPLIANT SWITCHES BY RYU CONTROLLER ON A MININET TESTBED

We present an approach for monitoring and measuring online per-flow as well as per-port packet loss statistics in SDN. The controller polls all the switches of the network periodically for port and flow statistics via OpenFlow 1.3 multipart messages. The OpenFlow compliant switches send cumulative statistics of sent and received packets to the controller that includes raw packets (control, non-user generated packets responsible for network management); which, although not being part of the end-to-end data traffic, get counted and act as noise in the statistics. The proposed method takes into account the effect of raw packets and thus, hamper the accuracy of methods.

BITS APOGEE 3D WEBSITE

TECH. FEST WEBSITE

The official website for BITS Pilani's annual technical fest - APOGEE. The first 3D website of any college fest in India. The website was made using three.js in frontend and Django in backend.

IMPROVING CROSS LANGUAGE TRANSLITERATION SYSTEM

TRANSLITERATION IN HINDI

Transliteration module provides way to transliterate from English to any Indian Language or from any Indian Language other Indian Language. This module is originally part of Swathanthra Indian Language Processing Applications. http://libindic.org/Transliteration

CAPTURE OF CARBON DIOXIDE FROM AMBIENT AIR

TRANSLITERATION IN HINDI

A Project for National Level CBSE Science Exhibition 2011, organised by CBSE held at Modern Public School, Barakhambha Road, New Delhi. The project was aimed at reducing the cost of Carbon Sequestration. It involved passing of ambient air through a channel where CO2 was adsorbed using ammonia based resin.

.05

CERTIFICATIONS

  • edX
  • SPRING
    2013

    FOUNDATIONS OF COMPUTER GRAPHICS

    UC BERKELEY

    Understood the concepts of 3D graphics, and developed programs that create images of a 3D scene with lighting, using both real-time OpenGL and GLSL programming, as well as offline raytracing.
  • FALL
    2012

    INTRODUCTION TO COMPUTER SCIENCE AND PROGRAMMING

    MIT

    Challenging and rigorous course to learn using computation to solve real problems.
  • Coursera
  • SEP
    2013

    ALGORITHMS: DESIGN AND ANALYSIS, PART 1

    STANFORD UNIVERSITY

    Rigorous course for "Big-oh" notation, sorting and searching, divide and conquer (master method, integer and matrix multiplication, closest pair), randomized algorithms (QuickSort, contraction algorithm for min cuts), data structures (heaps, balanced search trees, hash tables, bloom filters), graph primitives (applications of BFS and DFS, connectivity, shortest paths).
  • JAN
    2014

    MACHINE LEARNING

    STANFORD UNIVERSITY

    A broad introduction to machine learning, datamining, and statistical pattern recognition. Topics include Supervised learning (parametric/non-parametric algorithms, support vector machines, kernels, neural networks), Unsupervised learning (clustering, dimensionality reduction, recommender systems, deep learning), Best practices in machine learning (bias/variance theory; innovation process in machine learning and AI).
  • JULY
    2014

    THE DATA SCIENTIST’S TOOLBOX

    JOHNS HOPKINS UNIVERSITY

    An introduction to version control, markdown, git, GitHub, R, and RStudio
  • JULY
    2014

    R PROGRAMMING

    JOHNS HOPKINS UNIVERSITY

    Using R for effective data analysis
  • JULY
    2014

    GETTING AND CLEANING DATA

    JOHNS HOPKINS UNIVERSITY

    An introduction to data scrapping, data cleaning and data sharing
  • JULY
    2014

    EXPLORATORY DATA ANALYSIS

    JOHNS HOPKINS UNIVERSITY

    Covers the essential exploratory techniques for summarizing data and common multivariate statistical techniques
  • AUG
    2014

    REPRODUCIBLE RESEARCH

    JOHNS HOPKINS UNIVERSITY

    The course focuses on the concepts and tools behind reporting modern data analyses in a reproducible manner
  • AUG
    2014

    STATISTICAL INFERENCE

    JOHNS HOPKINS UNIVERSITY

    The course focuses on the process of drawing conclusions about populations or scientific truths from datausing statistical modeling, data oriented strategies and explicit use of designs and randomization
  • AUG
    2014

    REGRESSION MODELS

    JOHNS HOPKINS UNIVERSITY

    This course covers regression analysis, least squares and inference using regression models including special cases of the regression model, ANOVA and ANCOVA and analysis of residuals and variability
  • AUG
    2014

    PRACTICAL MACHINE LEARNING

    JOHNS HOPKINS UNIVERSITY

    The course focusses on concepts such as training and tests sets, overfitting, and error rates and introduces a range of model based and algorithmic machine learning methods including regression, classification trees, Naive Bayes, and random forests
.06

SKILLS

PROGRAMMING SKIILLS
Learning programming and computational skills since high school
LEVEL : ADVANCEDEXPERIENCE : 5 YEARS
C C++ Java Python MySQL R MATLAB OpenGL
WEB STACK
Have developed unique websites for college's sports, cultural and tech. fests. 3D website for Apogee 2014
LEVEL : ADVANCEDEXPERIENCE : 4 YEARS
HTML5|CSS3|JSNodejsdjangojQueryWebRTCthree.js
SOFTWARE DEFINED NETWORKS
Exploring this new field. Interested in hybrid networks and network monitoring
LEVEL : EXPLORINGEXPERIENCE : 1.5 YEARS
MininetOpen vSwitchOpenFlowRyu ControllerQuagga
DATA SCIENCE
Interested in character recognition, Among top 25% in various data science challenges at Kaggle
LEVEL : INTERMEDIATEEXPERIENCE : 2.5 YEARS
EnsembleNeural NetsComputer VisionImagination
DESIGN SKILLS
Designed websites for college fests, have shot and edited live events at Department of Visual Media, BITS Pilani
LEVEL : INTERMEDIATEEXPERIENCE : 3 YEARS
Adobe Photoshop CS6Adobe Premier Pro CS6Prezi
TOOLS
Various technical tools I am familiar with
LEVEL : INTERMEDIATEEXPERIENCE : VARIES
Git Visual StudioEclipseWP SDK8
MUSIC AND MISCELLANEOUS
Playing Tabla since age of 7. Have performed as vocalist, percussionist and dancer at national level youth festivals
LEVEL : EXPERTEXPERIENCE : 15 YEARS
TablaVocalsDrumsCongo
.07

GALLERY

img11

MISP, N. DELHI

img

Project title

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Fusce a auctor sem. Suspendisse egestas nulla eget nunc commodo, et blandit ante tristique. Aliquam dignissim nulla tellus, sed pellentesque libero pellentesque et. Donec nec sem mattis, suscipit ligula id, porttitor tortor. Maecenas sed egestas odio, vitae euismod nulla. Duis viverra blandit mi quis rhoncus. Aenean vitae turpis et tortor elementum blandit.

dignissim nulla tellus, sed pellentesque libero pellentesque et. Donec nec sem mattis, suscipit ligula id, porttitor tortor. Maecenas sed egestas odio, vitae euismod nulla. Duis viverra blandit mi quis rhoncus. Aenean vitae turpis et tortor elementum blandit.

img11

ON TELEVISION

img

Project title

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Fusce a auctor sem. Suspendisse egestas nulla eget nunc commodo, et blandit ante tristique. Aliquam dignissim nulla tellus, sed pellentesque libero pellentesque et. Donec nec sem mattis, suscipit ligula id, porttitor tortor. Maecenas sed egestas odio, vitae euismod nulla. Duis viverra blandit mi quis rhoncus. Aenean vitae turpis et tortor elementum blandit.

dignissim nulla tellus, sed pellentesque libero pellentesque et. Donec nec sem mattis, suscipit ligula id, porttitor tortor. Maecenas sed egestas odio, vitae euismod nulla. Duis viverra blandit mi quis rhoncus. Aenean vitae turpis et tortor elementum blandit.

img11

CLICKCOUNSELOR

img

Project title

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Fusce a auctor sem. Suspendisse egestas nulla eget nunc commodo, et blandit ante tristique. Aliquam dignissim nulla tellus, sed pellentesque libero pellentesque et. Donec nec sem mattis, suscipit ligula id, porttitor tortor. Maecenas sed egestas odio, vitae euismod nulla. Duis viverra blandit mi quis rhoncus. Aenean vitae turpis et tortor elementum blandit.

dignissim nulla tellus, sed pellentesque libero pellentesque et. Donec nec sem mattis, suscipit ligula id, porttitor tortor. Maecenas sed egestas odio, vitae euismod nulla. Duis viverra blandit mi quis rhoncus. Aenean vitae turpis et tortor elementum blandit.

img11

YOUTH FESTIVAL

img

Project title

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Fusce a auctor sem. Suspendisse egestas nulla eget nunc commodo, et blandit ante tristique. Aliquam dignissim nulla tellus, sed pellentesque libero pellentesque et. Donec nec sem mattis, suscipit ligula id, porttitor tortor. Maecenas sed egestas odio, vitae euismod nulla. Duis viverra blandit mi quis rhoncus. Aenean vitae turpis et tortor elementum blandit.

dignissim nulla tellus, sed pellentesque libero pellentesque et. Donec nec sem mattis, suscipit ligula id, porttitor tortor. Maecenas sed egestas odio, vitae euismod nulla. Duis viverra blandit mi quis rhoncus. Aenean vitae turpis et tortor elementum blandit.

.08

CONTACT

Get in touch

Interested in research collaboration and related
I would be happy to talk to you if you want to collaborate with me, or advise me, or need my assistance in research or music or anything else. You can find me at the lab: Centre for Excellence in Software Defined Networking, 6021, NAB, BITS Pilani.
mail.firstname.lastname@gmail.com