BTech in AI

BTech in AI


Artificial Intelligence (AI) is an area that spans multiple disciplines and impacts almost every sphere of life. It is widely agreed that AI has the potential to have a large impact on “the human condition”, making it imperative that engineers of tomorrow are trained in the scientific, engineering, as well as sociological aspects of AI. AI, by its nature, is a multidisciplinary subject, both in terms of the set of techniques needed to develop it as well as the set of fields it has interactions with. Broadly speaking, we can understand it as the study of strategies that harness data collection and computation in order to enhance and empower human capabilities. While statistical and computational techniques form the core of any AI subfield, the importance of domain knowledge cannot be underestimated. Our designed curriculum is based on the following principles:

1. It provides a solid foundation for the basics in AI, from both the computational as well as statistical perspective. While students do not need a view into all aspects of computer science, they need enough exposure to identify, e.g. what are the hard problems in AI and why, and what engineering is involved in an implementation of AI algorithms.
2. The curriculum is designed to be flexible to adapt to a rapidly changing field.
3. The curriculum is designed to give students enough leeway to specialize in any of the subfields of AI, i.e. students should be able to train themselves in the relevant AI techniques, choose a domain, and then use the AI techniques to synthesize novel questions and applications in their domain of interest.

Over the last few years, the number of faculty in different disciplines working on applications of AI in their respective domains has significantly increased. Familiarity with AI concepts, to various extents, is in high demand both in the industry as well in the government sector. It is therefore proposed that a major in Artificial Intelligence be established at the Institute.
1. Discipline Core Courses - 44 credits
2. Discipline Elective Courses - 20 credits. This consists of the following baskets.
    a. Advanced AI basket,
    b. AI application basket,
    c. CSE elective basket.

AI techniques to synthesize

Core Courses

The discipline-specific curriculum consists of the following core courses.

Course code Name Credits
ES 242 Data Structures and Algorithms -1 4
ES 203 Digital Systems 4
CS 301 Theory of Computing 4
ES 215 Computer Organization & Architecture 4
CS 2xx (to be proposed) AI Software Tools and Techniques 4
AI/CS 2xx (to be proposed) Mathematical Foundations of AI 4
ES 654 Machine Learning 4
To be proposed Signals, systems and random processes 4
EE xxx (Electrical) Control Systems 4
CS 328 Introduction to Data Science 4
To be proposed Foundations of Artificial Intelligence 4
Total 44

Elective structure

Electives are divided into the following three baskets. 20 credits of electives must be taken. The "advanced AI" and "AI cluster" baskets are just a guiding structure. There is no constraint on the minimum number of credits from these two baskets.

Advanced AI: These courses cover advanced AI topics that are applicable to a variety of domains. No minimum credit limit.
AI applications: These courses cover various AI application clusters, such as language, speech, vision, etc. No minimum credit limit.
CSE Electives (at least 4 credits have to be taken) These courses offer students background in core EECS topics that can help students apply AI for X or use X for AI. For example, if X = Software Engineering, then students wishing to study software engineering for AI or AI for software engineering will find such a course useful.

The following table lists the electives in each basket. This list of courses is expected to be dynamic. When any course gets approved, we should identify the elective basket for that course. In general, listing a course in one of these buckets as an elective will involve going through its curriculum in a discipline meeting.

Selected online courses and CS 299/399/499 courses can contribute at most 8 credits towards the discipline electives.

Basket Course code Course Name Credits
CSE Electives ES 214 Discrete Math 4
CSE Electives CS 327 Compilers 4
CSE Electives CS 301 Operating Systems 4
CSE Electives CS 433 Computer Networks 4
CSE Electives EE 411 Digital Signal Processing 4
CSE Electives CS 434 Software Engineering and Testing 4
CSE Electives CS 432 Databases 4
AI advanced AI 4xx Reinforcement Learning 4
AI advanced AI 4xx Game Theory 4
AI advanced ES 413 Deep learning 4
AI advanced ES 645 Optimization methods in ML 4
AI advanced ES 661 Probabilistic ML 4
AI advanced AI 4xx Knowledge representation and reasoning 4
AI advanced CS 4xx (to be proposed) AI and Ethics 4
AI applications CS 613 Natural language processing 4
AI applications EE 645 3D Computer Vision 4
AI applications ES 659 Computer Graphics 4
AI applications AI 4xx Speech Technology 4
AI applications ES 615 Nature Inspired Computing 4
AI applications ES 4xx Robotics 4
AI applications AI 4xx AI for Sustainability 4
AI applications AI 4xx AI for Earth Sciences 4
AI applications AI 4xx AI for Biological engineering 4