CTech Teaching Load Autumn 2017-18

ELL782 Computer Architecture (PG) (3-0-0) 3 credits

Dr. Sumantra Dutta Roy [Slot A]

ELL781 Software Fundamentals for Computer Technology (3-0-0) 3 credits

Dr. Sumeet Agarwal, Dr. Sumantra Dutta Roy [Slot C]

JRL301 Robotics Technology (3-0-0) 3 credits

Prof. S. K. Saha, Dr. Sumantra Dutta Roy, Prof. Kolin Paul, Prof. I. N. Kar

ELL780 Mathematical Foundations of Computer Technology (3-0-0) 3 credits

Prof. Suresh Chandra. Dr. Sumeet Agarwal (Coordinator) [Slot M]

ELP781 Digital Systems Lab (0-0-4) 2 credits
Subrat Kar [Slot P]

ELP718 Telecommunication Software Laboratory (0-1-4) 3 credits
Subrat Kar (Slot S)

ELL818 Telecom Technologies (3-0-0) 3 credits
Subrat Kar [Slot M]

ELL787 Embedded Systems and Applications (3-0-0) 3 credits
Dr. S. M. K. Rahman, Prof. Subrat Kar, Dr. Tapan K Gandhi [Slot X1]

ELL100 Introduction to Electrical Engineering (UG) (3-0-2) 4 credits

Prof. Jayadeva [Slot fixed]

ELL880 Special Topics in Computers-I (3-0-0) 3 credits

Social Network Analysis

Dr. Sougata Mukherjea & Dr. Amit A. Nanavati.

Dr. Sumantra Dutta Roy (Coordinator) [Slot X1]

[Fri 7:00 - 10:00 AM, EE Committee Room]

[Regular lectures: Tue and Fri, 07:30am-09:00am. Room LH-612.]

ELL881 Special Topics in Computers-II (3-0-0) 3 credits [Registration Cap: 40]

Practice of Deep Learning

Dr. Raghavendra Singh [Slot X1]

[Regular lectures: Thu 2-5pm. Room LH-603.]

ELL784 Introduction to Machine Learning (3-0-0) 3 credits

http://web.iitd.ac.in/~sumeet/ell784.html

Sumeet Agarwal [Slot J]

ELV780 Special Module in Computers (1-0-0) 1 credit

IoT: Internet of Things

Samsung Researchers. Dr. Sumantra Dutta Roy (Coordinator) [Slot X1]

[Introductory lecture: 02 Aug (Tue) 2pm-4pm. Venue: IIA-106, Bharti School (ground floor).

Regular lectures: 2pm-4pm. Dates: 9th Aug, 23rd Aug, 25 Aug, 13th Sept, 27th sept, 11th oct. (all Wednesday other than 25th Friday)

https://piazza.com/configure-classes/fall2017/elv780cov885 [piazza.com]

access code: elv780cov885

]

ELD880 Major Project Part-I (0-0-12) 6 credits

Dr. Sumantra Dutta Roy (Coordinator)

ELL793 Computer Vision (3-0-0) 3 credits

Dr. Brejesh Lall [Slot D]

ELL305 Computer Architecture (3-0-0) 3 credits

Dr. Tapan K Gandhi, Dr. Sumantra Dutta Roy, Dr. S. M. K. Rahman, Prof. Subrat Kar [Slot E]

ELL890 Computational Neuroscience (3-0-0) 3 credits

Dr. Tapan K Gandhi [Slot L]

ELL785 Communication Networks (3-0-0) 3 credits

Dr. Swades De [Slot X1]

ELV821 Selected Module in Communication Systems and Networking-II

Quality of Service (QoS) in Telecommunication Systems and Networks
Prof. H. M. Gupta [Slot X1]

ELV781 Special Module in Information Processing-I

Deep Reinforcement Learning

Dr. Ankur Narang (Senior VP, Tech & Decision Sciences, Yatra Online Pvt Ltd) [Slot X1]

[Course-related logistics: Dr. Narang plans to start classes in the week of 18 Sep. Possible timing 8-9am on Tue and Fri. He may add some Saturdays to substitute for weekdays. He may put a registration cap of 60. Since this is a course with manual registration (the course will start on 18 Sep), to trim the course registration, some suitable policy will be followed. An example of such a policy is here [link]. The final policy as suggested by the instructor (in conjunction with Departmental considerations) is as follows:

1. PhD students: EE/Bharti first, then others.
2. Masters students: EE/Bharti first, then Duals in their M.Tech phase.
3. BTech students: EE first, in order of seniority.

Class slot: Wed 7:30am-9am, LH-519 [room allotted: confirmed]

(The other alternatives were: LH-521/LH-604/LH-611/LH-613/LH-615/LH-422(tight fit))

The list of selected people (those selected: above the red line): [link]

Course Notes and Descriptions:

ELL881 Practice of Deep Learning

• Course Objective: Introduce students to the basic concepts underlying deep learning. The concepts will be reinforced by a thorough discussion of some key networks applied to computer vision and natural language processing. In this course we will demo applications of DL, introduce the biological motivation behind DL, cover basic theoretical aspects of         DL, learn packages that implement DL, so that at end of course each student can demo a novel application of DL.

• Key Modules:

• Basics: Cover the underlying theory of deep learning

• Introduction to Deep Learning
• Neural         Network Architecture, Activation, Loss function
• Optimization
• Programming packages for deep learning

• Concrete Networks: Each of these would be accompanied by a hands-on lab session, where students will implement the network using tensorflow in the class by seeking help of instructor. This would be concluded by a discussion on the best way to implement the network.

• Convolution Neural Networks (CNN) for Vision
• Fully Connected Network for NLP
• Recurrent Neural Network (RNN) for NLP
• Generative Models for Vision

• Grading        

• Lab Participation 15%, MidTerm 15%, FinalTerm 30%, Project 40%

• Lecturer(s): Raghavendra Singh, Vineet Kumar IBM
• Pre-requisites

• Python, Linear Algebra, Prob & Stats
• IN CLASS LAB SESSIONS         would be more beneficial to those that have access to a basic laptop on which tensorflow can be run using docker containers.                 

• Texts/Other Resources

• I. Goodfellow, A. Courville, Y. Bengio         Deep Learning
• CS231n: Convolutional Neural Networks for Visual Recognition
• CS224d: Deep Learning for Natural Language Processing

ELV 821 Selected Module in Communication Systems and Networking-II

Quality of Service (QoS) in Telecommunication Systems and Networks

Instructor: Prof H M Gupta (Email: hmgupta@ee.iitd.ac.in; harimgupta@gmail.com)

Expansion of Telecommunication Services is taking place in a rapid

fashion. The ultimate test is the acceptable Quality of Service

(QoS). The QoS  cannot be the  same for all services and a suitable

Framework needs to be specified meeting technical, operational and regulatory

requirements. This course is designed to look in to the above and

associated issues.

The course is designed for M.Tech.  ( Telecommunication Technology and

Management), M.Tech. (Communications). MS(R), Ph.D. students. M.Tech

students  from other streams can also credit this course.

Course timing: Flexible – dependent on students and instructor’s

convenience (1.5 Hr slot per week is preferred)

Proposed Contents:

QoS Background, Definition, Characteristics, Metrics and Parameters

QoS Management Schemes: Policy based, Integrated Services,

Differentiated services

QoS Architectures: End-to-end, Control, Technology Centric

QoS Mapping: Reference Architecture, Modules, Layering

QoS in Multimedia Communications: Voice, Video, Data

QoS in Cellular Networks (3G, LTE, 4G and beyond): Attributes for

Conversational, Streaming, Interactive and Background Classes

Measurement of QoS

Current Trends

References:

1.    Wiliam C Hardy :QoS Measurement and Evaluation of Telecommunications

Quality of Service, Wiley, 2001

2.    Mario Marchese: QoS over Heterogeneous Networks, Wiley, 2007

ELV780 Special Module in Computers (1-0-0) 1 credit

IoT: Internet of Things

Course Plan [Link]

Project Proposal [Link]

ELV781 Special Module in Information Processing-I

Deep Reinforcement Learning

• Course Objective: This course provides introduction to the concepts and applications of Deep Reinforcement learning (DRL). The course will start with overview of Reinforcement Learning and Deep Learning and then move on to DRL overview, key techniques used in solving DRL formulations         and its applications in multiple areas and industry verticals. The         students will be encouraged to learn packages that implement DRL, and evaluation will be based on demo of an interesting application of DRL.

Key Modules:

• Introduction to Reinforcement Learning & Deep Learning
• Concepts in Reinforcement Learning & Deep Learning
• Markov Decision Processes, Value & Policy Iterations
• Model         based and Model free learning, Q-learning
• Deep Networks: CNNs, Gradient Descent Methods
• Introduction & Concepts in Deep Reinforcement Learning
• Policy         Iteration, Actor-critic method
• Applications of Deep Reinforcement Learning
• Packages: OpenAI gym, rllab, rl.py
• Autonomous         robot motion
• Playing Computer Games such as Atari
• Business operations

Grading        

•         Project 100%

•         Lecturer(s): Ankur         Narang

Pre-requisites

• Python, Linear Algebra, Prob & Stats
• Machine Learning foundations including classification, regression algorithms
• Reinforcement learning familiarity

Texts/Other Resources

• CS 294: Deep Reinforcement Learning         (UC Berkeley)
• CS 234: Reinforcement Learning         (Stanford)        
• CMU 10703: Deep Reinforcement Learning & Control         
• Sutton & Barto, Reinforcement Learning: An Introduction
• Goodfellow, Bengio & Courville, Deep Learning