number, advertisement, yellow, colour, asphalt, road, digit, numbering, two, number, number, number, number, number

BITS WILP MTech AIML – Semester 2 – Knowledge Base

/ AIML - Artificial Intelligence & Machine Learning, SEM 2 - AIML / By

Orientation & Electives for BITS WILP AIML – Sem 2

Before start of the semester two of BITS WILP AIML there be a batch wide orientation session in which the students are introduced to various subjects and electives structure and how to choose each. Each subject will have a prescribed handout which lists the topics and syllabus for that course.

Refer here to choose electives for WILP AIML (as per orientation), here for slides of sem-2 orientation.

  • Two courses are mandatory (Core Courses) & the other two electives can be chosen by students
  • It is not compulsory to choose a specialization, you can choose any electives and it wont have any impact on your degree/graduation. Even if you choose random electives AIML it is fine. It is preferrable to choose based on your domain & interest.
  • After the AIML sem-2 orientation they will send out a survey and based on the survey results they will decide which electives to keep and plan for faculty and sections (This is only for operational planning from BITS faculty to decide on courses faculty etc.). But final electives choosing will happen once the results for semester one are published
  • During registration there will be details on class schedule, faculty etc. Sections & Electives once selected during registration cannot be changed.

Sem 2 – Core & Electives for WILP AIML

📚Core Courses (Mandatory)
Click on each of the below for more details

Deep Reinforcement Learning – Overview
  • This is core course, hence mandatory
  • Deep Reinforcement Learning Handout – here
  • Detailed course about the basic concept of Reinforcement learning that was introduced in semester. This is like a core course like that of ACI or ML, will act as a foundation for various future courses.
  • Contents: Lots of algorithms related to reinforcement learning. Mathematical foundations of deep reinforcement learning
  • Webinars are python based, Assignments are mostly programmatic (python) based but also has some reasoning questions
  • Exams are numerical focused

Deep Neural Networks – Overview
  • Deep Neural Networks Handout – here
  • This is core course, hence mandatory
  • Concepts: Fundamental concepts of deep learning concepts & deep neural networks, Artificial neural networks, evaluation & optimization of deep learning models
  • Pre-requisites: Concepts learned in semester 1, backward and forward propagation, basic ML math
  • This course will have opportunities to develop the neural networks from scratch using PyTorch & Tensorflow
  • One research oriented assignment, two programmatic assignments using available libraries
  • Exams will have a mix of numerical, justifications, reasoning questions

📚Elective Courses (To Choose Any Two)
Click on each of the below for more details

Natural Language Processing – Overview
  • Natural Language Processing Handout – here (Mandatory elective for NLP specialization)
  • Concepts covered: Fundamentals of language processing, acts as a foundation for LLM related concepts. Main topics are Language modelling, Part-of-speech tagging (HMM), Topic modelling, Vector semantics, Knowledge graphs & basic RAG architecture etc.
  • Pre-requisites: A bit of math will be required, probability, basic understanding of sigmoid SoftMax functions
  • Assignments will mostly be group assignments

Information Retrieval – Overview
  • Information Retrieval Handout – here (Preferrable for NLP specialization)
  • Covers topics related to structure & organization of various components of an IR system; Data retrieval models for a search system; Recommender systems; Understand architecture of search engines, web crawlers etc.
  • No pre-requisites for this course, you can take this course even if you don’t know anything about this.
  • Good to take if you wish to understand RAGs in the future.
  • This course is NOT math intensive, basic mathematics is enough

Probabilistic Graphical Methods – Overview
  • Probabilistic Graphical Methods Handout – here (General Elective)
  • Course is about encoding probability distributions over large numbers of random variables that interact with each other.
  • Major concepts covered will be Probability theory + Graph models (Directed & Undirected Graph models)
  • Pre-requisites: Probability theory, bayes rule, probability density functions, cumulative distribution functions. It will be a theoretical course (theorem + proof learning & exams will be conceptual). Math intensive course. If you enjoy MFML you will enjoy this course

Data Management For Machine Learning – Overview
  • Data Management For Machine Learning Handout – here (General Elective)
  • Concepts covered: Deals with various data technology Topics & underlying concepts from Data ingestion such as Kafka, Data structuring like Spark, Python, Scala, Data orchestration like Airflow, DBT, Aggregation tools like Snowflake, BI tools like tableau; Architecture of modern data stack, Various data phases
  • The course has hands-on demonstrations & expert sessions through webinars
  • No Pre-requisites: If you have some basics of databases its good, but not compulsory as long as you keep an open mind to learn.

Fair Interpretable Transparent Machine Learning – Overview
  • Fair Accountable Transparent Machine Learning Handout – here (Preferrable for Deep Learning specialization)
  • Concepts covered: Related to fairness, accountability, transparency & interpretability of ML models; You will be building ML models in the course and interpret
  • Heavy programmatic assignments using python
  • Exams are mostly numerical in nature with some theory/application questions

ML System Optimization – Overview
  • ML System Optimization handout – here (Preferrable for Deep Learning specialization)
  • Concepts: Related to hardware optimization, GPUs & its architecture, GPU optimization, CUDA architecture, optimizing some of the existing ML algos like KNN to ensemble techniques. Focus a lot on back propagation algos and techniques to parallelize
  • Assignments are Programmatic based and Research oriented, Assignments may require the knowledge of C or C++ & algorithm engineering.
  • Pre-requisites: Concepts of ML, Basics of C or C++ (Not too intense), Basics of hardware architecture preferred (ALU working)
  • No standard textbooks, you will have to go through PPT and Research papers; Overall subject is theoretical and algorithmic based

AIML Techniques For Cyber Security – Overview
  • AIML Techniques For Cyber Security handout – here (General Elective)
  • Concepts: Apply ML techniques to cyber security, ML for anomaly detection, Malware detection, Network intrusion, Profiling network traffic
  • Pre-requisites: A bit of python or any language for ML, basics of cyber security concepts
  • Assignments are programmatic based – they give a framework code and you have to build on that
  • Exams are a mix of both theory & numericals

WILP AIML – Tentative Academic Calendar

Second semester starts from May (This will change depending on which cohort you’ve joined in). Mid-semester exams in July and the semester concludes in September.

📅 Exam Time Table – 2025

Exam Type Exam Phase Dates
Mid-Semester (EC2) Regular 27th, 28th, 29th June, 2025
Make-up 11th, 12th, 13th July, 2025
Comprehensive (EC3) Regular 5th, 6th, 7th September, 2025
Make-up 12th, 13th, 14th September, 2025

Note: A detailed exam schedule will be shared in a separate mail.

BITS WILP AIML – Subject Wise Resources

📚AIML-ZG530: Natural Language Processing
Click on each of the below for more details

NLP Previous Year Papers
  • 2024-08 – NLP EC2 (Mid Sem Exam Regular) Paper – here
  • 2024-08 – NLP EC2 (Mid Sem Exam Makeup) Paper – here
  • 2024-01 – NLP EC2 (Mid Sem Exam Regular) Paper – here
  • 2020-01 – NLP EC2 (Mid Sem Exam Makeup) Paper – here
  • 2020-01 – NLP EC2 (Mid Sem Exam Regular) Paper – here
  • 2020-03 – NLP EC3 (End Sem Exam Regular) Paper – here
  • 2020-03 – NLP EC3 (End Sem Exam Makeup) Paper – here
NLP Sample Papers & Question Banks
  • Compilation – Solved Problems Question Bank – here
  • Compilation – Question Bank – here
  • Compilation – Previous Papers Bank – here

NLP Other Important Links & Resources
  • Topic Wise Important Learning Resources (BITS Recommended) – here

NLP Contact Session Slides – Topic Wise
  • Contact Session 1 – Introduction here
  • Contact Session 2 – Vector Semantics & Embeddings here
  • Contact Session 3 – Word Embeddings here
  • Contact Session 4 – Language Modelling here
  • Contact Session 5 – Neural Networks & Neural Language Modelling here
  • Contact Session 6 – POS & HMM here
  • Contact Session 8 – Revision For Mid Sem here
  • Contact Session 9 – Grammars & Parsing here
  • Contact Session 10 – PCFG here
  • Contact Session 11 – Dependency Parsing here
  • Contact Session 12 – Contextual Embedding here
  • Contact Session 13 – Word Senses & Word Net here
  • Contact Session 14 – Semantic Ontology here
  • Contact Session 15 – Text Summarization here
  • Contact Session 16 – Revision For Final Sem here

📚AIML-ZG511: Deep Neural Networks
Click on each of the below for more details

Deep Neural Networks – Previous Year Question Papers
Deep Neural Networks – Research Papers
  • Continual Lifelong Learning with Neural Networks A Review – Download PDF
  • Federated Learning Of Out Of Vocabulary Words – Download PDF
  • Federated Machine Learning Concept and Applications – Download PDF
  • INTRODUCTION TO NEURAL NETWORKS – Download PDF
  • Meta Learning in Neural Networks A Survey – Download PDF
  • NEURAL NETWORKS AS UNIVERSAL APPROXIMATORS – Download PDF
  • NIPS 2012 imagenet classification with deep convolutional neural networks Paper – Download PDF
  • Three scenarios for continual learning – Download PDF
Deep Neural Networks – Other Resources

📚AIML-ZG512: Deep Reinforcement Learning
Click on each of the below for more details

Deep Reinforcement Learning – Previous Year Question Papers
Deep Reinforcement Learning – Research Papers
  • A Comprehensive Survey on Safe Reinforcement Learning – Download PDF
  • BAIL Best Action Imitation Learning for Batch Deep Reinforcement Learning – Download PDF
  • Deep Reinforcement Learning with Double Q learning – Download PDF
  • Efficient Architecture Search by Network Transformation – Download PDF
  • Human level control through deep reinforcement – Download PDF
  • Imitation Learning A Survey of Learning Methods – Download PDF
  • Large Scale Evolution of Image Classifiers – Download PDF
  • Learning Latent Dynamics for Planning from Pixels – Download PDF
  • Mastering Atari, Go, Chess and Shogi by Planning with a Learned Model – Download PDF
  • Mastering the game of Go without human knowledge – Download PDF
  • Rainbow Combining Improvements in Deep Reinforcement Learning – Download PDF

📚AIML-ZG537: Information Retrieval
Click on each of the below for more details

Information Retrieval – Previous Year Question Papers

📚AIML-GENS: Data Management For Machine Learning
Click on each of the below for more details

Data Management For ML – Previous Year Question Papers
  • DSECLZG529 AIMLCZG529 Data Management for Machine Learning Compre_Makeup AK – Download PDF
  • DSECLZG529 AIMLCZG529 Data Management for Machine Learning Compre_Regular AK – Download PDF
  • DSECLZG529 AIMLCZG529 Data Management for Machine Learning Midsem_Makeup AK – Download PDF
  • DSECLZG529 AIMLCZG529 Data Management for Machine Learning Midsem_Regular AK 2 – Download PDF

🎓 Start Here – BITS WILP – AIML (Artificial Intelligence & Machine Learning)
Shopping Cart