EXCLUSIVE: Amazon may expand ML Summer School to neighbouring countries

To upskill machine learning (ML) talent in the country, Amazon piloted ML Summer School for students in Indian universities last year. The e-commerce giant is into the second edition of the ML Summer School that will focus deep into different ML topics with supervised learning, unsupervised learning, probabilistic graphical models, deep learning and more. Unique to India, Amazon may roll this ML Summer School to neighbouring countries, where the time zones aren’t very different, Rajeev Rastogi, Vice President - International Machine Learning at Amazon told Business Today.

“This ML Summer School has evolved from a similar, internal program ‘ML Gurukul’ that we have been running for a few years to upskill Amazonians internally. The idea is to make thousands of students industry ready for science roles every year. This ML Summer School covers a mix of both theoretical and practical industry knowledge. We are looking to train and prepare the students so that they can solve practical problems in the industry. Multiple organizations within Amazon are involved in rolling it out - AWS and also the retail part of the business,” says Rastogi.

In the first year of our ML Summer School for 2021, Amazon shortlisted 317 students to participate from 20 selected institutes of India for this skill development program. The company reached out to the Indian Institute of Technology (IIT) in its Bombay, Madras, Kharagpur, Delhi and Kanpur campuses, along with several other leading institutes, like the Indian Institute of Science in Bengaluru and many more.

“For the second year, we have opened our ML Summer School to students in the penultimate or final year of their Bachelor’s/Master’s/Integrated Masters/PhD degree enrolled in any campus in India. The top 3000 engineering students will be enrolled in ML Summer School - who attend eight virtual classroom sessions over four weekends in July,” adds Rastogi.

Last year’s module included Supervised Learning, Deep Neural Networks, Probabilistic Graphical Models, Dimensionality Reduction and Unsupervised Learning. However, this year, two new modules - Reinforcement Learning and Causal Inference - have been added.