Free Course Image Artificial Intelligence Masterclass: Search, Logic, Machine Learning and Deep Learning

Free online courseArtificial Intelligence Masterclass: Search, Logic, Machine Learning and Deep Learning

Duration of the online course: 31 hours and 57 minutes

New

Free NPTEL AI masterclass covering search, logic, planning, probabilistic reasoning, machine learning, deep learning, NLP, ethics, and GenAI.

In this free course, learn about

  • Course Orientation and Foundations of AI
  • Agents and Problem Solving
  • Core Search Strategies
  • Heuristic and Informed Search
  • Adversarial Search and Game Playing
  • Local Search and Evolutionary Computation
  • Logic, Knowledge, and Reasoning Agents
  • Predicate Logic and Inference
  • Expert Systems
  • Planning and Scheduling
  • Knowledge Representation and Non-Monotonic Reasoning
  • Representation Schemes
  • Reasoning Under Uncertainty and Probabilistic Models
  • Utility and Decision Theory
  • Introduction to Machine Learning
  • Ensemble Methods and Core Neural Network Concepts
  • Hypothesis Learning and Practical MLP Work
  • Deep Learning Foundations and Datasets
  • CNNs and RNNs with Practical Work
  • Generative AI and Fuzzy Logic
  • Natural Language Processing and Responsible AI
  • Swarm and Multi-Agent Systems

Course Description

Artificial Intelligence Masterclass: Search, Logic, Machine Learning and Deep Learning is a free online course from NPTEL designed for learners who want a structured path from foundational AI concepts to modern deep learning and generative AI. It fits the Technology and Programming category and focuses on Artificial Intelligence and Machine Learning with a strong balance of theory, reasoning, and practical perspectives.

Start by building core intuition around agents and environments, problem solving, and search strategies, including uninformed and informed search as well as adversarial game-playing approaches. From there, progress into evolutionary computation with genetic algorithms and genetic programming, connecting optimization ideas to intelligent decision-making.

The course then develops a rigorous logic and knowledge representation foundation, covering propositional logic, predicate logic, and inference in first-order settings. You will also explore knowledge-base agents, expert systems, and key planning concepts such as planning graphs, scheduling, and practical approaches for representing and reasoning about the world.

To handle real-world uncertainty, the curriculum introduces probabilistic reasoning through Bayesian networks and dynamic belief networks, alongside utility and decision theory. It then transitions into machine learning essentials, including learning algorithms, ensemble methods, hypothesis learning, and neural network fundamentals such as multilayer perceptrons.

Finally, dive into deep learning workflows with datasets, CNNs, RNNs, and practical sessions that connect models to implementation thinking. The course extends to natural language processing, ethics in AI, generative AI, fuzzy sets with fuzzification and defuzzification methods, and advanced collective intelligence topics like swarm intelligence and multi-agent systems.

Course content

  • Video class: AI Concepts and Techniques Intro 08m
  • Exercise: Which topic is introduced as the starting point of the course overview?
  • Video class: Lec 01 Course overview 30m
  • Exercise: Which set of capabilities was listed as necessary for a computer system to pass the Turing test?
  • Video class: Lec 02 History 32m
  • Exercise: What key limitation made early neural networks less effective before hidden layers became popular again in the 1980s?
  • Video class: Lec 03 Agents and Environment 35m
  • Exercise: Which statement best defines a rational agent?
  • Video class: Lec 04 General Problem Solving 34m
  • Exercise: In a search-based problem formulation, which component specifies the successor state reached by applying an action in a given state?
  • Video class: Lec 05 Introduction to Search Strategies 28m
  • Exercise: In graph search (compared to tree search), what additional check prevents generating redundant paths?
  • Video class: Lec 06 Uninformed Search Strategies - I 26m
  • Exercise: In Breadth-First Search (BFS), which node is selected for expansion at each step?
  • Video class: Lec 07 Uninformed Search Strategies - II 31m
  • Exercise: What is the key idea behind Iterative Deepening Depth-First Search (IDDFS)?
  • Video class: Lec 08 Informed Search Strategies - I 29m
  • Exercise: In the A* (A-star) search algorithm, how is the evaluation function f(n) defined?
  • Video class: Lec 09 Informed Search Strategies - II 29m
  • Exercise: In memory-bounded A* (e.g., SMA*), what action is taken when available memory is exhausted during search?
  • Video class: Lec 10 Adversarial or Game-Playing Search 31m
  • Exercise: In the minimax procedure for two-player zero-sum games, how are values backed up from children to a parent node?
  • Video class: Lec 11 Game-Playing Search 29m
  • Exercise: In alpha-beta pruning, what do the parameters α (alpha) and β (beta) represent?
  • Video class: Lec 12 Introduction to Genetic Algorithms 31m
  • Exercise: In local search, which statement best describes what matters when moving through the solution space?
  • Video class: Lec 13 GA Operators 32m
  • Exercise: In roulette-wheel (stochastic sampling with replacement) selection in a genetic algorithm, how is the chance of a string being chosen determined?
  • Video class: Lec 14 Genetic Programming 29m
  • Exercise: In genetic programming, how is crossover typically performed on program representations?
  • Video class: Lec 15 Knowledge-Base Agents 30m
  • Exercise: In a knowledge-based agent, which pair best describes what is stored in the knowledge base (KB)?
  • Video class: Lec 16 Proposotional Logic 32m
  • Exercise: In propositional logic, which statement correctly describes the connective “X implies Y” (X → Y)?
  • Video class: Lec 17 Predicate Logic - 1 30m
  • Exercise: Which statement best captures a key advantage of predicate (first-order) logic over propositional logic?
  • Video class: Lec 18 Predicate Logic - 2 32m
  • Exercise: In first-order logic for the family-relationships domain, how is a grandparent relationship correctly characterized?
  • Video class: Lec 19 Inference in Predicate Logic 33m
  • Exercise: In predicate-logic resolution, what does deriving nil (empty clause) after adding the negation of the goal to the knowledge base indicate?
  • Video class: Lec 20 Inference in First Order Logic 33m
  • Exercise: Which technique speeds up forward chaining by only considering rules where at least one premise fact was newly inferred in the previous iteration?
  • Video class: Lec 21 Introduction to Expert Systems 31m
  • Exercise: In an expert system, what is the primary role of the inference engine?
  • Video class: Lec 22 Applications of Expert Systems 20m
  • Exercise: In an expert system, which components are typically kept separate so that different inference methods can operate on the same knowledge?
  • Video class: Lec 23 Introduction to Planning 31m
  • Exercise: In planning using PDDL, what is typically included in an action schema?
  • Video class: Lec 24 Planning Graph 40m
  • Exercise: In a planning graph, when is the graph typically terminated (stopped from further expansion)?
  • Video class: Lec 25 Planning and Scheduling 27m
  • Exercise: In scheduling after planning, what identifies the critical path in a plan?
  • Video class: Lec 26 Knowledge Representation 29m
  • Exercise: In knowledge representation using categories, what does it mean when two categories are disjoint?
  • Video class: Lec 27 Non Monotonic Reasoning 26m
  • Exercise: In non-monotonic reasoning, what does a Truth Maintenance System (TMS) help an agent do when a fact is retracted?
  • Video class: Lec 28 Representation Schemes - 1 32m
  • Exercise: In a semantic network, how does the inheritance algorithm typically determine a property value (e.g., number of legs) for a specific object?
  • Video class: Lec 29 Representation Schemes - 2 30m
  • Exercise: In a conceptual graph, what do the circular/elliptical nodes represent?
  • Video class: Lec 30 Uncertanity in Knowledge Representation 38m
  • Exercise: In a decision-theoretic agent, how is the final action typically chosen under uncertainty?
  • Video class: Lec 31 Bayesian Network 26m
  • Exercise: In a Bayesian network, what does a directed arrow from node X to node Y represent?
  • Video class: Lec 32 Dynamic belief networks (DBN) 33m
  • Exercise: In a first-order Markov process used in a Dynamic Bayesian Network, which dependency is assumed for the transition model?
  • Video class: Lec 33 Utility Theory 31m
  • Exercise: How does a rational agent choose an action according to the Maximum Expected Utility (MEU) principle?
  • Video class: Lec 34 Decision Theory - 1 28m
  • Exercise: In multiattribute decision making, what does it mean if option S1 has strict dominance over option S2?
  • Video class: Lec 35 Decision Theory - 2 30m
  • Exercise: Which set of components correctly defines a Markov Decision Process (MDP)?
  • Video class: Lec 36 Introduction to Learning 34m
  • Exercise: Which option correctly matches a learning type with the kind of feedback it uses?
  • Video class: Lec 37 Learning Algorithms 35m
  • Exercise: In the K-nearest neighbors (KNN) classifier, how is a new pattern P typically assigned a class label?
  • Video class: Lec 38 Learning Algorithms 31m
  • Exercise: When constructing a decision tree, how is the attribute for a node typically chosen?
  • Video class: Lec 39 Ensemble Learning 25m
  • Exercise: In ensemble classification, what does the simplest and most robust method use to combine outputs from multiple classifiers?
  • Video class: Lec 40 Introduction to Neural Networks 30m
  • Exercise: In training an artificial neural network, what is primarily updated to make the network produce the desired output for a given input?
  • Video class: Lec 41 Multilayer Perceptron Neural network 19m
  • Video class: Lec 42 Hypothesis Learning 43m
  • Exercise: In current best hypothesis learning, what update is performed when a new training example is a false negative for the current hypothesis?
  • Video class: Lec 43 Practical session - 1 42m
  • Exercise: In a multiclass classification problem with 3 classes, what output-layer setup is appropriate for an MLP?
  • Video class: Lec 44 Practical session - 2 58m
  • Exercise: In a binary classification neural network, which combination is appropriate for the output layer and loss function?
  • Video class: Lec 45 Introduction to deep learning 32m
  • Exercise: In deep learning, what is a key advantage regarding feature engineering compared to standard machine learning?
  • Video class: Lec 46 Datasets for DL 30m
  • Exercise: Which dataset is described as having 14 million labeled images spanning more than 2,000 categories?
  • Video class: Lec 47 CNN and introduction to RNN 41m
  • Exercise: In a CNN, what is the main purpose of pooling (e.g., max pooling or average pooling) after convolution?
  • Video class: Lec 48 RNN 34m
  • Exercise: In a conditional GAN (cGAN), what additional input is provided to both the generator and the discriminator to enable targeted generation?
  • Video class: Lec 49 Practical session - 3 1h08m
  • Exercise: In image representation for CNNs, how many channels are typically used for an RGB (color) image?
  • Video class: Lec 50 Practical Section - 4 25m
  • Exercise: In a 4-class CNN image classification task using Keras, which loss function and output activation are most appropriate?
  • Video class: Lec 51 Practical Section - 5 29m
  • Exercise: In a many-to-one RNN setup for stock prediction, what does the model output represent?
  • Video class: Lec 52 Generative AI 36m
  • Video class: Lec 53 Fuzzy Sets and Systems 29m
  • Exercise: In fuzzy set theory, how is the union of two fuzzy sets A and B typically computed for an element x?
  • Video class: Lec 54 Fuzzification and Defuzzification methods 34m
  • Exercise: In fuzzy set theory, what does the core of a membership function represent?
  • Video class: Lec 55 Natural Language Processing 32m
  • Exercise: Which type of ambiguity occurs when a single word has multiple meanings (e.g., bank as a financial institution or river edge)?
  • Video class: Lec 56 Ethics in AI 39m
  • Exercise: Which set of principles is highlighted as key aspects of AI ethics?
  • Video class: Lec 57 Swarm Intelligence 35m
  • Exercise: In swarm intelligence, which statement best explains how complex global behavior arises?
  • Video class: Lec 58 Multi agent systems 34m
  • Exercise: In a multi-agent system, which description best matches negotiation between agents?

This free course includes:

31 hours and 57 minutes of online video course

Digital certificate of course completion (Free)

Exercises to train your knowledge

100% free, from content to certificate

Ready to get started?Download the app and get started today.

Install the app now

to access the course
Icon representing technology and business courses

Over 5,000 free courses

Programming, English, Digital Marketing and much more! Learn whatever you want, for free.

Calendar icon with target representing study planning

Study plan with AI

Our app's Artificial Intelligence can create a study schedule for the course you choose.

Professional icon representing career and business

From zero to professional success

Improve your resume with our free Certificate and then use our Artificial Intelligence to find your dream job.

You can also use the QR Code or the links below.

QR Code - Download Cursa - Online Courses

More free courses at Artificial Intelligence and Machine Learning

Free Ebook + Audiobooks! Learn by listening or reading!

Download the App now to have access to + 5000 free courses, exercises, certificates and lots of content without paying anything!

  • 100% free online courses from start to finish

    Thousands of online courses in video, ebooks and audiobooks.

  • More than 60 thousand free exercises

    To test your knowledge during online courses

  • Valid free Digital Certificate with QR Code

    Generated directly from your cell phone's photo gallery and sent to your email

Cursa app on the ebook screen, the video course screen and the course exercises screen, plus the course completion certificate