Programme
Industrial Engineering and Operations Research at IIT Bombay
(IEOR@IITBombay) and McKinsey Knowledge Centre India have partnered to
create a challenging competition for students interested in Analytics.
Three winners of the competition stand to win an iPad Pro, Bose wireless
headphones, and a Kindle Oasis. The winners of this competition will also
have an opportunity to directly present their findings during the much
awaited IEOR Day @ IIT Bombay.
Ronald Coase had stated "If you torture the data long enough, it will
confess". We hope that this competition will ignite the innovator in you
and give you a glimpse of how data analytics techniques are useful in
solving real life operations research problems.
Registration for the competition has started. Contest starts from 10th
February and ends on 19th February.
To register and participate click on the button below
Topic: Machine Larning in Banking and Financial Services
Dr Aniruddha Pant, CEO-Founder, AlgoAnalytics and co-founder, A2IoT, has completed his PhD in control systems from University of California, Berkeley. He has 20+ years of experience in application of advanced mathematical techniques to various domains. His expertise spans ?nancial engineering, quant-trading, hedging, machine learning and control theory.
We talk about challenges faced by BFSI companies like retaining customers, improving sales, cross-sell and how our ML/AI solutions enable companies to overcome those by using churn modeling, recommender systems, summarization techniques, sentiment analytics among others. He will also include how machine learning is helping banks to automate tasks like signature verifcation, automated email responses etc.
Ajit is co-founder of two start ups SmartCommute (Uber for corporate employee transport) and DeepTek (Radiology AI). Previously Ajit was the CEO co-founder of Vertex Software - an IT services company focused on Japanese market. Vertex Software had strategic equity partnership with Mitsui & Co. ltd. - top Japanese conglomerate and exited to NTTDATA - Japan's leading IT company. Ajit is a graduate of IIT Kharagpur.
Ankit has over twelve years of experience across Financial Services and Management Consulting including roles in Risk Governance, Strategy and Program Management. Ankit is currently working as Vice President Risk Management at Nomura as well as the Fintech Innovation Lead for Risk Management. He has previously worked for Accenture Strategy and KPMG. Ankit Garg holds an MBA from Thunderbird School of Global Management at the Arizona State University.
Curious about Analytics, and Passionate in refining the interface between Analytics and Users* - Graduated from Pune University in Hotel Management Technology - Over 10 years of experience with Hotel Operations and Hotel Revenue Management Technology - Hotel Operations Experience with Hyatt Bombay. 4 years - 7 Years with IDeaS Revenue Solution at various roles. As a Revenue Analyst, Client Relationship Manager for EMEA, Solution Specialist with engineering solutions for Pilots and New Markets, to now being the Manager for Client Experience - primarily involved in recruiting and grooming talent towards ensuring a seamless experience for the user while using IDeaS Analytical Solutions - When not working ? Photography, Travel and Family
Having leadership expertise in all vertical of Data Science of delivering end to end solution using different tools/technique with experience in Data Analysis & Modeling, Data warehousing and BI/Analytics & Reporting, Sanchit Tiwari is currently Practice Expert at McKinsey & Company.
He has delivered multiple analytical models using different Data Mining and Machine learning algorithm as Classification, Clustering, Regression, Decision trees, Asociation rule mining, SVM, Random Forest,Ensemble methods, Neural Network, Deep learning. Along with having domain experience with data science in Retail( B2C, B2B), Finance, HR Analytics with expertise on customer analytics.
Experienced in end to end automation of data science projects and Expertise in data warehousing & building end to end data models, he has working knowledge of SAS ETL and BI tools like SAS Data Integration Studio & SAS Web Report Studio. SAS Enterprise Guide/Miner. He also has basic knowledge of other ETL tool like Informatica.
Seshadri is the founder and CEO of Spashta Technologies, a company that is
building the next generation of solutions for supply chain decision makers. He
brings to the table extensive experience in supply chain management. He began his
career at i2 Technologies, where he specialized in process consulting, solution
implementation, pre-sales, and training. He worked on an array of projects for
several Fortune 500 companies across the FMCG, CE&CD, Automotive, Industrial, and
Metal sectors. Seshadri also gained expertise at Dell for 7 years, where he played
multiple roles like he designed planning solutions for the BTS supply chain, was in
charge of replenishment and solution development for finished goods in APJ, and
was also a program manager for implementation of Oracle VCP solutions.
With a B.Tech. from IIT(Bombay) and an M.S (IE&OR) from the University of
Cincinnati, Seshadri understands supply chains from a global perspective. His
cross-functional experience spanning nearly two decades has sharpened his ability
to understand and develop accurate solutions for every business problem.
Topic: Blockchain & Machine Learning
Mechanism design (MD) provides a game theoretic framework to explore if a given social
choice function may be implemented as an equilibrium outcome of an induced game. In a
multi-agent setting, machine learning (ML) seeks to learn the preferences or types of the
agents through any available data or through intelligent exploration.
ML and MD are well investigated as individual problems, however, interesting research questions arise when both of them are required for solving a problem. Many current problems involving strategic agents holding private information (for example, Internet advertising auctions, Crowdsourcing, smart grids, smart contracts with blockchains, and online educational platforms) need MD and ML to be used together. In this talk, we examine some technical challenges that arise in solving such problems. In particular, we focus on our current research into design of multiarmed bandit mechanisms where online learning and mechanism design blend together to yield a powerful new modeling option for many important problems in AI.
Y. Narahari is currently a Professor at the Department of Computer Science and Automation, Indian Institute of Science, Bangalore, India. The focus of his research in the last decade has been to explore problems at the interface of computer science and game theory. He has recently authored a textbook entitled “Game Theory and Mechanism Design” brought out by the IISc Press and the World Scientific Publishing Company. He is a Fellow of the IEEE and a J.C. Bose Fellow of the Department of Science and Technology, Government of India. He is a Fellow of INSA, IASc, NASI, and INAE. More details at: http://lcm.csa.iisc.ernet.in/hari/. The focus of my current research is to apply game theory, mechanism design, and machine learning to current research problems at the interface of computer
Blockchains: An Overview
Dr. Praphul Chandra is the Founder of koinearth - a stealth mode startup working at the intersection of Mechanism Design, Blockchains and Machine Learning. He is also a professor of data science and machine learning at the International School of Engineering (Insofe).Prior to this, he was Principal Data Scientist at Hewlett Packard Enterprise where his focus was on the application of machine learning techniques to solve real world problems across multiple domains like the Internet of Things, Taxation Fraud, Telecom and Social Network Analytics. His other industry experience includes positions at HP Labs and Texas Instruments. He has an undergraduate degree in Electronics engineering from IIT BHU, a post graduate degree in Electrical Engineering from Columbia University, NY, a post graduate diploma in public policy from University of London and a PhD in Game Theory & Mechanism Design from the Indian Institute of Science.
The Internet of Value. The Future of Money. The Trust Machine. A Universal Computer. Blockchain has been called many things by many people. It is a little like the fable of five blind men and the elephant. Perhaps no one really understands what all blockchain is and yet there there is an excitement about the possibilities that it opens up. To some, it is reminiscent of the early days of the world wide web with grand visions of how it can change the world.
This talk will seek to give a brief overview of blockchains. The ?rst application of blockchains was the bitcoin cryptocurrency. Even though bitcoin is remarkable in what it has achieved, it is the underlying blockchain technology that has potential applications beyond currency. The most ambitious applications are about re-imagining today's social institutions.
At its heart, what blockchain achieves is decentralized trust. Using some very innovative application of cryptography and distributed computing, it allows nodes in a peer to peer network to achieve a consensus. In monetary applications, this consensus is about the transactions that have taken place and are recorded in a distributed ledger. Placed in the larger social context however, money itself is just a tool in an accounting system of trust and reputation - the two pillars of modern social life. Thus, what blockchain enables is the opportunity to re-engineer social institutions.
We are however, only at the beginning of this journey. Blockchain, as a technology presents both opportunity and challenges. For example, one of the key challenges is of being able to scale blockchain technology; another one is to prevent misuse by antisocial elements. But the opportunities far outweigh the challenges and the applications are limited only by one's imagination. And finally, lets not forget the huge data set that blockchains make publicly available.
A Visionary & Engaging leader having 10 years of experience in Supply Chain Analytics and Continuous Improvement leading a team of 25 analysts. He developed a high-performance advanced analytics team from 5 to 60 members in 7 years to deliver differential value creation and competitive advantage & identified $200+ million savings opportunities for North American business and expanded to South America, Europe, Middle East and Asia. He has broad experience in building new teams, new capabilities, expanding teams, mobilize top-tier talent to create a high-performance culture with the high sense of accountability. His interest lies in Supply Chain, Project management, Executive Presentation and Reporting Descriptive analytics, Diagnostic analytics, and Data Visualization as well as Leading Data Science projects in Supply Chain domain.
Currently, he is the Professor and Dean at School of Technology and Computer Science, TIFR. From December 1996 to December 2002 he was a faculty in the Operations Research Group in the Department of Mechanical Engineering at IIT Delhi. He has also held visiting positions at many places including at Columbia University, Stanford University and Indian School of Business. He was then also a member of the Bank of America's executive quantitative council. He is currently on the editorial board of Stochastic Systems. Earlier he has been on editorial boards of Mathematics of Operations Research, Management Science, and ACM TOMACS.
He spent a sabbatical and then was an adjunct at Centre for Advanced Financial Research and Learning (CAFRAL), a research wing of Reserve Bank of India (2015-16).
His research interests lie in applied probability including in mathematical finance, Monte Carlo methods, and game-theoretic analysis of queues.
Anirudh Patil is the director for India Knowledge Centers Strategy & Operations who leads the McKinsey India Knowledge Centers in Gurugram and Chennai, helps in providing industry and functional expertise, advanced analytics, business research, and proprietary tools and data. He joined the McKinsey Singapore office in 2004 as a consultant where he served clients in the high-tech and telecommunications industries. In early 2013, he joined the McKinsey India Knowledge Center (McKC) seeing an exciting opportunity to drive change and stand at the cutting edge of innovation for the firm.
Anirudh has worked closely with his colleagues to create an open culture and promote cross collaboration across every aspect of McKC's work. Recently, he introduced three new "spider" teams that help incubate new capabilities, develop new leadership talent, promote diversity, and encourage greater external outreach throughout McKC. He is positioning McKC to play an integral role at the forefront of the firm's innovative knowledge and analytical capabilities. He wants to foster a spirit of bold experimentation and identify and develop high-potential talent, even as he ensures that McKC remains cost effective.
He hold a degree in MBA from INSEAD along with PGDM in Marketing from Xavier Institute of Management and Entrepreneurship and a BS degree in Engineering from University of Mumbai.
Raman Sankaran is a Research Scientist working with Conduent labs, Bangalore. His primary research interests are in machine learning, more specifically towards optimization using structured sparse regularizers. Other research interests include application of machine learning for transportation problems, time series modeling and multi-task learning. Prior to joining Conduent, He worked with Prof. Chiranjib Bhattacharyya in IISc, Bangalore as a PhD scholar.
Tushar Shekhar is currently Senior Project Manager, Product Development at JDA Software. He holds B. Tech in Industrial Engineering from IIT Kharagpur and a degree in General Management for IT Executives from Indian Institute of Management, Banglore.
Topic: Analytics and Science for Growth and Consumer Experience at Flipkart
This talk will draw heavily from the Big Data and Analytics related Initiatives and Actual use-cases in transport sector, especially Indian Railways
Milind Gullavani is a Director supply chain analytics with Flipkart. He has around 14 years of experience in the field of Supply Chain, Analytics and Product Management. He has done my M.Tech in Industrial Engineering. Before Flipkart, he has worked with Amazon, Philips and Mercedes Benz. In Flipkart, he manages E2E speed and reliability to make sure to have good customer experience.
Topic: Collaborative Supply Chains using Blockchain
The blockchain technology is expected to transform the way the supply chains are organized and the way collaboration takes place on inter-organizational supply chain tasks. At IBM Research, we have been conducting several interesting in-market experiments in using the blockchain technology for Supply Chain Collaboration. In this talk, we will present an on-the-field-view of how such projects are conceived and what it takes to execute such projects on the ground! We will also provide an overview of the research agenda that we are pursuing in this broad space.
Dr. Vinayaka Pandit is the Senior Manager of Blockchain Research at IBM India Research Labs. In this role, he directs the research that the department conducts in the systems and applications side of the Blockchain technology. Vinayaka has spent last 19 years in IBM Research, starting out as a researcher working in approximation algorithms. He worked extensively on applying ideas from Approximation Algorithms and Combinatorial Optimization to several research areas within IBM Research. Before taking on the role of leading Blockchain Research, Vinayaka led the Mathematical Modeling and Algorithmic Solutions research group which specialized in taking techniques from Algorithms, Optimization, and Machine Learning to challenging real world problems in traditional industries like Oil & Gas, Metals & Mining.
We are interested in performance evaluation of a stochastic classifier (governed by a posterior distribution), in terms of bounds on its average true risk obtained via PAC-Bayesian theorem. Our aim is to choose an informative posterior relevant to these bounds. A major contribution is the use of these results to compute PAC-Bayesian bound based intervals for averaged true risk when the regularization parameter in SVM design is chosen as per given prior and posterior distributions.
Simulation is one of the effective and popular tool in modeling such complex stochastic models based on their behavioral characteristics and its input parameter settings. The simulation models are executed under different input parameter settings and its corresponding output performances are captured. The critical decisions on real system are made based on the output performance of the simulation model. Most simulation models are used for performance evaluation of real systems, comparing alternatives and finding the best/optimal configurations etc. Simulation-based Optimization (SbO) assumes that the simulation model is valid, and that the probability distributions used therein are accurate.However, in practice, the input probability distributions (input models) are estimated by sampling data from the real system. The errors in such estimates can have a profound impact on the optimal solution obtained by SbO. To tackle above problem we consider the robust simulation based optimization problem. we present an algorithmic approaches for solving the robust simulation based optimization problem. Our algorithmic procedure is based on the stochastic kriging metamodel-assisted bootstrapping with an efficient global optimization technique which sequentially searches the optimum.
We derive a closed form description of the convex hull of mixed-integer bilinear covering set with bounds on the integer variables. This convex hull description is completely determined by considering some orthogonal disjunctive sets defined in a certain way. Our description does not introduce any new variables. We also derive a linear time separation algorithm for finding the facet defining inequalities of this convex hull. We show the effectiveness of the new inequalities using some examples. For the unbounded case, we identify the split-rank one and higher rank facet defining inequalities. We also find the conditions under which the optimal objective value over the convex hull of the set is same as that over all the rank-one facet defining inequalities. A relaxation of the trim loss problem has this property, where as our computational results show that for trim loss problem, only rank one inequalities are sufficient.
In a bus transportation system the time gap between two successive buses is called headway. When the headways are small (high-frequency bus routes), any perturbation (e.g., in the number of passengers using the facility, traffic conditions, etc.) makes the system unstable, and the headway variance tends to increase along the route. Eventually, buses end up bunching, i.e, they start traveling together. Bus bunching results in an inefficient and unreliable bus service and is one of the critical problems faced by bus agencies. Another important aspect is the expected time that a typical passenger has to wait before the arrival of its bus. The bunching phenomenon might reduce if one increases the headway, however this can result in unacceptable waiting times for the passengers. We precisely study this inherent trade-off and derive a bus schedule optimal for a joint cost which is a convex combination of the two performance measures.
In many security and healthcare systems, a sequence of sensors/tests is used for detection and diagnosis. Each test outputs a prediction of the latent state, and carries with it inherent costs. Our objective is to learn strategies for selecting a test that gives the best trade-off between accuracy & costs for given context. Unfortunately, it is often impossible to acquire ground truth annotations and we are left with the problem of unsupervised sensor selection (USS). With contextual information, we propose an algorithm having sub-linear regret and verify our results on synthetic & real datasets.
Topic: Machine Larning in Banking and Financial Services
Dr Aniruddha Pant, CEO-Founder, AlgoAnalytics and co-founder, A2IoT, has completed his PhD in control systems from University of California, Berkeley. He has 20+ years of experience in application of advanced mathematical techniques to various domains. His expertise spans ?nancial engineering, quant-trading, hedging, machine learning and control theory.
We talk about challenges faced by BFSI companies like retaining customers, improving sales, cross-sell and how our ML/AI solutions enable companies to overcome those by using churn modeling, recommender systems, summarization techniques, sentiment analytics among others. He will also include how machine learning is helping banks to automate tasks like signature verifcation, automated email responses etc.
Ajit is co-founder of two start ups SmartCommute (Uber for corporate employee transport) and DeepTek (Radiology AI). Previously Ajit was the CEO co-founder of Vertex Software - an IT services company focused on Japanese market. Vertex Software had strategic equity partnership with Mitsui & Co. ltd. - top Japanese conglomerate and exited to NTTDATA - Japan's leading IT company. Ajit is a graduate of IIT Kharagpur.
Ankit has over twelve years of experience across Financial Services and Management Consulting including roles in Risk Governance, Strategy and Program Management. Ankit is currently working as Vice President Risk Management at Nomura as well as the Fintech Innovation Lead for Risk Management. He has previously worked for Accenture Strategy and KPMG. Ankit Garg holds an MBA from Thunderbird School of Global Management at the Arizona State University.
Curious about Analytics, and Passionate in refining the interface between Analytics and Users* - Graduated from Pune University in Hotel Management Technology - Over 10 years of experience with Hotel Operations and Hotel Revenue Management Technology - Hotel Operations Experience with Hyatt Bombay. 4 years - 7 Years with IDeaS Revenue Solution at various roles. As a Revenue Analyst, Client Relationship Manager for EMEA, Solution Specialist with engineering solutions for Pilots and New Markets, to now being the Manager for Client Experience - primarily involved in recruiting and grooming talent towards ensuring a seamless experience for the user while using IDeaS Analytical Solutions - When not working ? Photography, Travel and Family
Seshadri is the founder and CEO of Spashta Technologies, a company that is
building the next generation of solutions for supply chain decision makers. He
brings to the table extensive experience in supply chain management. He began his
career at i2 Technologies, where he specialized in process consulting, solution
implementation, pre-sales, and training. He worked on an array of projects for
several Fortune 500 companies across the FMCG, CE&CD, Automotive, Industrial, and
Metal sectors. Seshadri also gained expertise at Dell for 7 years, where he played
multiple roles and he designed planning solutions for the BTS supply chain, was in
charge of replenishment and solution development for finished goods in APJ, and
was also a program manager for implementation of Oracle VCP solutions.
With a B.Tech. from IIT (Bombay) and an MS (IE&OR) from the University of
Cincinnati, Seshadri understands supply chains from a global perspective. His
cross-functional experience spanning nearly two decades has sharpened his ability
to understand and develop accurate solutions for every business problem.