C.Tech Courses for the Spring Semester 2018-19 (II Semester, 2018-19)
(based on the C.Tech Group Meeting dated 03 Oct 2018, and subsequent discussions)
- ELL100 Introduction to Electrical Engineering
Subrat Kar, Tapan Kumar Gandhi
- ELL888 - Advanced Machine Learning (Neural networks and Deep Learning) [Registration Cap: 60]
A P Prathosh [Slot M]
- [MTech CTech Core] ELL783 Operating Systems (3-0-2) 4 credits
Sumantra Dutta Roy [Slot A]
- ELL784 Introduction to Machine Learning (3-0-0) 3 credits [Registration Cap: 50]
Sumantra Dutta Roy [Slot C]
- [BTech Dept Core] ELL365 Embedded Systems (3-0-0) 3 credits
S. M. K. Rahman (Coordinator) [Slot A]
Instructors: Subrat Kar, S. M. K. Rahman
- ELL409 Machine Intelligence and Learning (3-0-2) 4 credits [Registration Cap: 50]
Jayadeva, A P Prathosh [Slot J]
- [Core] ELP305 Design and System Lab (0-0-3) 1.5 credit
S. M. K. Rahman (co-ordinator) [Slot R]
Instructors:
Prathosh A. P. (1 day of the week)
S. M. K. Rahman (1 day of the week)
Seshan Srirangarajan (1 day of the week)
Subrat Kar (1 day of the week)
Tapan Kumar Gandhi (1 day of the week)
- ELL457 Special Topics in Cognitive and Intelligent Systems (3-0-0) 3 credits
Sumeet Agarwal, Sumitava Mukherjee (HSS) [Slot M]
- ELL880 Special Topics in Computers-I (3-0-0) 3 credits
Computational Learning Theory and the Mind
Sumeet Agarwal [Slot X1]
- ELV780 Special Module in Computers (1-0-0) 1 credit
“IoT: Internet of Things”
Instructors: Samsung researchers [Slot X2: mutually convenient timings]
Sumantra Dutta Roy (Coordinator)
- ELL402 Computer Communication (3-0-0) 3 credits
Swades De
[Slot X3: timings: mutual convenience]
-ELP782 Computer Networks Lab (0-1-4)
Swades De
[Slot P]
-ELL822 Special Topics in Communication Systems and Networking-II (3-0-0) 3 credits [A Comm Course]
Linear and Nonlinear Programming
Jun-Bae Seo
[Slot X4]
-ELL896: Mobile Computing (3-0-0) 3 credits
[Slot: X5: timings: mutual convenience]
H M Gupta (harimgupta@gmail.com; hmgupta@ee.iitd.ac.in) (Coordinator: Swades De)
- ELD880 Major Project Part-1 (Computer Technology) (0-0-12) 6 credits
Sumeet Agarwal [Slot P]
- ELD881 Major Project Part-2 (Computer Technology) (0-0-24) 12 credits
Sumeet Agarwal [Slot P]
- ELD780 Minor Project (Computer Technology) (0-0-4) 2 credits
Sumeet Agarwal [Slot P]
- ELV781 Special Module in Information Processing-I (0-0-1) 1 credit
“Machine Learning and Economics”
Instructors: Raghavendra Singh (IBM-IRL), Geeta Singh (Genesis Analytics)
Sumantra Dutta Roy (Coordinator)
Notes:
Course Outlines:
ELV781 Special Module in Information Processing-1
Machine Learning and Economics
Economics is the study of allocation of scarce resources. Conceptually the allocation decisions made by consumers, firms or governments reflect the best outcome that these economic agents can achieve given the constraints they face. Economic theory models this decision making process as a complex optimization exercises under one or more constraints, making mathematical modeling an integral part of rigorous economic analysis. However, the relevance and applicability of many of these economic models becomes clear only when they are taken to data. With the explosion in the volume of data that is becoming available, development of tools to effectively examine and use this data is critical. Although statistical theory is the foundation of all data analysis, advances in field-specific empirical tools have pushed boundaries of empirical analysis as in other fields.
In this spirit, this course examines the intersection of economics and machine learning. After a brief introduction to economic modeling, the course will cover the basic data analysis tools used in economics as well as those used in machine learning. The objective is to examine how ML tools can be used to improve the empirical analysis of economic data. This course will focus on current work by Klienberg, Athey et.al. that is bringing these two fields closer.
The application of these econometric and ML tools for prediction and for estimating causal effects will be emphasized. For example, portfolio managers are interested in predicting future returns on past stock returns. On the other hand, the ability to estimate the causal effect of a past government intervention such as reducing the cost of education, or predict its effect in the future is important for policy makers.
READING
Text
Joshua Angrist and Jörn-Steffen Pischke. Mostly Harmless Econometrics: An Empiricist's Companion.
Articles
Hal R. Varian. Causal inference in economics and marketing. PNAS, July 5 2016
Susan Athey and Guido Imbens. The Econometrics of Randomized Experiments, arXiv July 2016
Kleinberg, Jon, Jens Ludwig, Sendhil Mullainathan, and Ziad Obermeyer. Prediction Policy Problems. American Economic Review, May 2015
Susan Athey. Beyond prediction: Using big data for policy problems. SCIENCE, Feb 2017
Susan Athey et. al. Matrix Completion Methods for Causal Panel Data Models, arXiv September 2018
Dubb, Jean-Pierre & Misra, Sanjog. Scalable Price Targeting, SSRN Electronic Journal 2017. 10.2139/ssrn.2992257.
A.Belloni, V.Chernozhukov, and C.Hansen. High-dimensional methods and inference on structural and treatment effects, Journal of Economic Perspectives,28(2):29–50,2014
COURSE EVALUATION
Students will be evaluated on the basis of an empirical project in economics or finance.
PreReq
Familiarity with Machine Learning, Python, Linear Algebra, Probability
Interested students may directly get in touch with Dr. Raghavendra Singh at:
raghavsi.vfaculty@ee.iitd.ac.in
From Dr. Singh:
I will start by introducing some ML basics basically part 1 of deep learning
book (about 3 hours).
https://www.deeplearningbook.org/
Then my fellow instructor will introduce problems and datasets in
Econometrics (about 4 hours).
Then we will do a machine learning and econometrics followup along the lines
of https://www.aeaweb.org/conference/cont-ed/2018-webcasts (Machine
learning and Econometrics section) (about 4 hours).
And then come back to discuss papers and datasets in this area (some of
which are listed in the handout).
I am a PhD from USC. I am senior researcher in IBM Research working in
computer vision field. I have taught a few classes in IIT delhi before on
deep learning etc. I am interested in understanding what are the other areas
that machine learning could be applied to.
Geeta Singh is a PhD in Economics from Stanford. She is working with an
economic consulting firm. She has taught before in Law Schools and IIITD.
She is interested in understanding how machine learning could be applied to
econometrics problems.
On the class schedule, from Dr.Singh:
“Based on the poll we are going to start on January 15th. The split on 2 vs 1
classes per week was 50% but let us try to make it two classes of 1 (<1.30)
hours per week. Classes will be held on Tues-Thurs afternoon, 5pm. Will send a mail with
Location. (I don’t want to do a poll on the class timings because Tues-Thurs afternoon
works best for us. You can send us mail if there is an issue.)”
Please consider LH615 to be the chosen location, unless informed otherwise.
ELL822 Special Topics in Communications (Optimisation)
Note from Jun-Bae Seo:
This course is about optimization techniques; the syllabus of this course is
1. Convex sets and functions
2. Constrained minimization theory: Lagrangian multiplier, KKT condition, Duality
3. Linear programming
4. Unconstrained minimization: Line search, Steepest descent, Newton, Quasi-Newton method and convergence; subgradient method for non-differentiable functions
5. Constrained minimization algorithms: Active set method, Penalty and Barrier method, Feasible direction method, Sequential Quadratic programming, Projected Gradient method
6. Applications to computer networks: Network utility maximization, and network decomposition
The main aim of this course is to help students get familiar with optimization so that they feel comfortable in reading research papers, when students would do their own research.
ELP782: Computer Networks Lab, 0-1-4, 3 Credits
This is the last call for registration for ELP782 (0-1-4, 3-credits) in 2017-18, Semester II. With the current number of registrations the course will not run in this semester.
Course objective: Intended to provide the protocol implementation concepts of communication network systems, and to be able to appreciate the connection between theoretical analysis and actual network system-level implementations.
Course content: Simulation and hardware experiments on different aspects of computer communication networks. Network traffic generation and analysis, differentiated service queues, network of queues using discrete event simulations.
Lab activities:
* Network traffic generation and analysis of traces; numerical verification using traffic distributions
* Learning discrete event simulation tools
* Simple point-to-point communication and queueing analysis
* Differentiated traffic generation, reception, and analysis
* Simulation of network of queues - open as well as closed network systems
* Exploration of application-specific network performance analysis
Self-study component:
* Cross-layer protocol optimization concepts: Distributed control, cost and energy efficiency
* Advanced research papers on case-by-case basis, per assignment
Tentative Course Evaluation Plan:
* Regular evaluation viva (30%)
* Regular evaluation demonstration (30%)
* Term project demonstration and viva (40%)
ELV780 Internet of Things - IoT Eco-system & Specifications
Link to course schedule: [link]
The course has a Piazza link (the code is elv780): http://piazza.com/iit_delhi/spring2019/elv780
whatsapp group for course: https://chat.whatsapp.com/BDA1dFxDtIE5eMSwO7rYLY
The lead instructor can be contacted at dinesh1.k [at] samsung [dot] com
ELL896: Mobile Computing
Audience: The course is open to postgraduate (M.Tech) students in Computer Technology, Computer Science and engineering Communication Engineering, Telecom Technolgy and Management, interested MS(R) and Ph.D. students, and senior B.Tech students.
Pre-requisite: None. Exposure to computer networking concepts would be useful.
The course will discuss essential concepts in Mobile Computing. These include:
Origination, philosophy and conceptualization. Issues and challenges
Physical layer infrastructure, wide area and local area architectures. Generations and Transitions – 5G issues and architecture, multicellular and heterogeneous networks
Networking protocol issues; mobility management, resource (bandwidth, energy)
aware MAC, network and transport protocols in mobile domain, opportunist multiconnect, data management and disconnected operation; location management, location based services
Portability, QoS issues, security and privacy
Recent trends: Mobile cloud computing, Integration with cloud storage, Anticipatory
mobile computing, context-awareness, Mobile Internet of Things (M-IoTs)
The course material will be drawn from relatively recent publications in professional
journals. The essential reading list will be given to students as the course progresses