Artificial Intelligence and Machine Learning at School:
This is an innovative program ( conceptualised, designed and developed by Prof. MM Pant to prepare a new generation of youth, who while still at School are initiated to the knowledge, skills, and growth mindset needed to flourish in the coming 4th Industrial Age driven by Artificial Intelligence and Machine Learning. It builds on the experience of and the soul of an “IIT education”, and inspired by Mahamana Madan Mohan Malaviya’s views on introducing technical education at an early stage while seeking to undo the immense damage done by McCaulay’s influence of creating a nation of clerks. This is elaborated in the last paragraphs.
We pilot with an introductory 10 hour program for learners at class 11 of Schools. The program is not limited to only Science stream students, but is desirable for all students, because AI is going to impact all areas of human activities.
The course is structured as 12 sessions of about 50 minutes each, and can therefore be delivered flexibly from an intensive 2 week course to a relaxed 3 months course, depending upon the learner needs.
Our default model is to run it as a Whatsapp course that can be accessed from a mobile phone ( Android or iOS) anywhere. The course will run for 2 weeks ( Monday to Saturday) with 5 posts per day of content that can each be transacted in about 10 minutes. This is nano-learning.
The time at which the posts will be made every day are after School hours at 4:30 pm, 5:00pm,5:30pm,6pm and 6:30 pm.
Two batches will run every month. One beginning on the first Monday of the month and the other beginning on the third Monday of the month. This is a tentative plan, and may be tweaked during the delivery for better effectiveness. The program has been conceived of, designed by and will be primarily delivered by Prof. MM Pant. He may draw upon the help and support of some other team members as needed.
It is our plan to be able to introduce a suitably modified version of this program to children at class 9 by the summer of 2019.
Tentative Session titles and session descriptors:
Session 1: The evolution of AI to its present state: what can AI do today?
Session brief : While AI seems as if it has suddenly appeared a few years ago, it has a long history. The term was first proposed in 1956 at a conference in Dartmouth. The big hopes led to some disillusionment and there was a ‘winter of AI’ with a Lighthill report that no further Government investments be made in the field. We identify the factors that have led to the great recent success of AI and the future hope from it.
The March of AI applications continues unabated, and is having profound applications in all sectors of the economy, with profound impacts for agriculture, healthcare, hospitality and education.
Topic 1.1@4:30pm: The early years
Topic 1.2@5:00pm: The winter of AI
Topic 1.3@5:30pm: The resurrection of AI
Topic 1.4@6:00pm: Drivers of the resurgence of AI
Topic 1.5@6:30pm: What can AI do today?
Session 2: Why School students should learn about AI,ML and CT?
Session brief: Most of the educational programs in the field are at Masters level and a few at the Undergraduate level. But we think that an appreciation and understanding of the field should be created at the School level itself. We have chosen Grade 11 as the stage at which we begin this initiative. All children at this stage have been born in the 21st Century and will be immersed in a world of AI technologies. Another important effect is raising questions about being a human?
Topic 2.1@4:30pm: What International political leaders are saying?
Topic 2.2@5:00pm: What Technology and business leaders are saying?
Topic 2.3@5:30pm: Why learning early matters?
Topic 2.4@6:00pm: Overcoming learning resistance
Topic 2.5@6:30pm: How it helps in making better choices for further education?
Session 3: Computational Thinking: Meaning,definitions and importance in the context of AI
Session brief : Jeanette Wing describes Computational Thinking as automation of abstraction. We consider this and other views. Then explain the elements of Computational Thinking: decomposition, pattern recognition, abstraction, algorithms and evaluation.
While the first conceptualisation of Computational Thinking was before the significant development of AI, but the extension was natural as pattern recognition is a common feature. Also understanding of algorithms that learn.
Topic 3.1@4:30pm: Computational Thinking: Meaning and definitions
Topic 3.2@5:00pm: Computational Thinking : in the context of Artificial Intelligence and Machine Learning
Topic 3.3@5:30pm: Algorithms
Topic 3.4@6:00pm: Learning Algorithms
Topic 3.5@6:30pm: Heuristics
Session 4: Relationship between Artificial Intelligence, Machine Learning and Deep Learning
Session brief: The 3 terms are used almost synonymously, but there are distinctions between these terms. In this session we explain that Machine Learning is a subset of Artificial Intelligence and Deep Learning is a subset of Machine Learning.
Topic 4.1@4:30pm: The Landscape of Artificial Intelligence
Topic 4.2@5:00pm: Machine Learning
Topic 4.3@5:30pm: Deep Learning
Topic 4.4@6:00pm: What are Artificial Neural Networks?
Topic 4.5@6:30pm: What problems are Artificial Neural Networks most suitable for?
Session 5: Artificial Neural Networks
Session brief: A big breakthrough in Machine Learning was when the analogy to neural networks was adopted to solve complex problems. Using suitable mathematical functions to represent the nodes(neurons) in the network with weighting factors gives a generalised model for solving many problems.
Topic 5.1@4:30pm: Feed forward neural network
Topic 5.2@5:00pm: Recursive Neural Networks
Topic 5.3@5:30pm: Recurrent Neural Networks
Topic 5.4@6:00pm: Convolutional Neural Networks
Topic 5.5@6:30 pm: An overview of other ANNs.
Session 6: Mathematics for AI and Machine Learning
Session brief : Many concepts learned in School Mathematics when extended and generalised find wide use in Machine Learning. Linear Algebra, Statistics and Calculus all find applications in implementing Machine Learning models.
Topic 6.1@4:30 pm: Where is the Mathematics used?
Topic 6.2@5:00 pm: Linear Algebra
Topic 6.3@5:30 pm: Calculus
Topic 6.4@6:00 pm: Probability and Statistics
Topic 6.5@6:30 pm: Optimisation techniques
Session 7: Object Recognition and Computer Vision
Session brief: The most well-known problem in computer vision consists of classifying an image into one of many different categories. In recent years classification models have surpassed human performance and it has been considered practically solved. However there are still plenty of challenges to image classification. Automated object recognition — and more generally scene analysis — from photographs and videos is the grand challenge of computer vision. This course presents the image, object, and scene models, as well as the methods and algorithms, used today to address this challenge.
Topic 7.1@4:30pm: Basic Principles of Computer vision: the foundations
Topic 7.2@5:00pm: What does it mean to have vision?Object tracking and localisation
Topic 7.3@5:30pm: How do Machines recognise objects? Possible approaches
Topic 7.4@6:00pm: Stages of an image classifier
Topic 7.5@6:30 pm: Convolutional Neural Networks for visual recognition
Session 8: Autonomous Transportation: How does it work
Session brief : Self-driving technologies have been developed by Google, Uber, Tesla, Nissan, and others.
Most self-driving systems create and maintain an internal map of their surroundings, based on a wide array of sensors. Software then processes those inputs, plots a path, and sends instructions to the vehicle’s “actuators,” which control acceleration, braking, and steering.
Obstacle avoidance algorithms, predictive modeling, and “smart” object discrimination help the software follow traffic rules and navigate obstacles.
Topic 8.1@4:30pm: The technical challenges to driverless cars
Topic 8.2@5:00pm: How they work? The technology leaders
Topic 8.3@5:30pm: The six Levels of driving Automation
Topic 8.4@6:00pm: Machine learning applied to autonomous transportation
Topic 8.5@6:30 pm: Impact that autonomous transportation will have
Session 9: Robots, Drones and humanoids
Session brief: Robots can be autonomous or semi-autonomous and range from humanoids to industrial robots, medical operating robots, patient assist robots, dog therapy robots, collectively programmed swarm robots and even microscopic nano robots. Robots traditionally were designed for performing specific tasks. Drones are robots that fly. But AI empowered robots can learn from experience.
Topic 9.1@4:30pm: Traditional robots
Topic 9.2@5:00pm: AI and Robotics
Topic 9.3@5:30pm: Drones
Topic 9.4@6:00pm: Swarm Robotics
Topic 9.5@6:30 pm: Humanoids
Session 10: Speech Recognition and conversational interfaces; Chatbots and Machine Translation
Session brief : The earliest user interface was a command line. It moved to GUIs in the recent past. The closer we get to a natural human interface, the more comfortable we will be solving problems. And the human natural interface is spoken language. It’s one of the first interfaces we ever came up with. And it has always been our favorite kind of interface.
Chatbots are algorithmic conversational agents which companies are coming up with to interact with their customers. The first stage in a chatbot is parsing the “utterance” and derive the “intention”. After generating the response using deep neural networks, it then again converts to human language.
Topic 10.1@4:30pm: What is a conversational interface?
Topic 10.2@5:00pm: The Challenges in Speech Recognition?
Topic 10.3@5:30pm: Natural Language Processing
Topic 10.4@6:00pm: Chatbots
Topic 10.5@6:30pm: Machine Translation
Session 11: The Technologies from IBM Watson, Google Tensorflow, Amazon and Apple
Session brief: IBM has Watson that has established itself with many applications. Google has Tensorflow and Duo and AlohaGo. Amazon has Alexa, Sagemaker and Deeplens congratulations big soon. Microsoft has its suite of tools. And Apple has its neural engine in its mobile phones that make them capable of Face recognition.
Topic 11.1@4:30pm: IBM Watson
Topic 11.2@5:00pm: Google Tensorflow
Topic 11.3@5:30pm: Microsoft
Topic 11.4@6:00pm: Apple
Topic 11.5@6:30pm: Amazon
Session 12: The Implications of AI: social, ethical and regulatory issues
Session brief : Several movies have painted a dismal picture of the effects of AI. Many reports have projected loss of jobs in large numbers. But the opportunities are also many. Our response has to be one of readiness and preparation for the new opportunities ahead and hence the need for an early orientation and awareness, at the School stage itself.
Topic 12.1@4:30pm: The disappearing jobs
Topic 12.2@5:00pm: The new job opportunities
Topic 12.3@5:30pm: The Gig economy
Topic 12.4@6:00pm: The Ethics of AI
Topic 12.5@6:30pm: Learning more about AI
Fee structure :
Fee for the 2 week program : Rs 5000+ GST Rs 900= Rs 5900
Fees can be easily paid by PayTM to mobile number :+919810073724
For those who would rather pay by Bank Transfer, here are the details:
For NEFT / RTGS
|Name||MADAN MOHAN PANT|
|Bank Name||HDFC BANK|
|Branch||Unit No.05 A & B, Ground Floor Tower A, Unitech Cyber Park, Sec 39,Gurgaon – 122002 Haryana|
|IFSC code||HDFC0002645 ( first four digits are alphabets and remaining 7 are numbers )|
For International remittances
|Name||MADAN MOHAN PANT|
|Bank Name||HDFC BANK|
|Branch||Unit No.05 A & B, Ground Floor Tower A, Unitech Cyber Park, Sec 39,Gurgaon – 122002 Haryana|
Customising the program:
At the request of educational Institutions ( Schools), this program can be customised for blended delivering with face to face sessions and Whatsapp sessions.
For any further queries or requests regarding this course, please send an e-mail to email@example.com : Or a Whatsapp message to Prof MM Pant at +919810073724
Motivation/ Purpose of this program:
I, Professor MM Pant am closely associated with the IIT system, which the Prime Minister Narendra Modi lauded recently at the IIT Mumbai convocation. I have received a Ph.D. in Computational Physics from IIT Roorkee ( it was called Roorkee University, when I joined the Ph.D. program there). I have taught as a faculty at IIT Kanpur from 1972 to 1980, and have been a member of the Board of Management of IIT Delhi for 6 years.
I have been a Professor of Computer Science at IGNOU ( and a Pro Vice-Chancellor very briefly) for about 15 years.
My mission therefore is to deliver a leading world class education but made available to a large number of the young.
It is now almost forgotten, but in the early years of IIT, a student was admitted after having studied class 11 and studied at IIT for 5 years to graduate as a B.Tech. We now have admission after passing class 12 with 4 years at IIT. Both add up to 16 years of Schooling.
The reason I am referring to this is that I am attempting to introduce the latest technologies to class 11 students, to orient them to the new ways of self-directed learning . By the time they complete a Bachelor’s degree they will be well equipped to face the emerging workplace with AI everywhere.
Lord McCauley felt in the 1830s, a need to produce – by English-language higher education -” a class of persons, Indian in blood and colour, but English in taste, in opinions, in morals and in intellect”, in other words clerks and low level bureaucrats. This was similar to the establishment of the Thompson College of Engineering in 1847 ….
But almost a hundred years ago in 1919, our Bharat Ratna Madan Mohan Malaviya had these thoughts on starting technical education at an early stage.
“Even if we begin to-morrow the Technical Education of all the youth of twelve years of age who have received sound elementary education, it will take seven years before these young men can commence the practical business of life, and then they will form but an insignificant minority in an uneducated mass.”
Around the Independence Day, 71 years after achieving our Independence, we are creating a movement for empowering the youth to liberate themselves from the stranglehold of the education regulators to become prosperous and flourish in the 4th Industrial Age.
This model has several innovative features: the first being to become future ready in knowledge and skills, learning at their place and time in short chunks of content (nano-learning) on their mobiles following personalised pathways. This is also an example of ‘consensual education’ where learners join courses that they want to pursue in the sequence that best suits them, rather than an authority driven bundle of mandatory courses. The age of Intelligent Machines will need ingenious humans to fully benefit from them, and this program leads them on that path.