IIT Ropar and NPTEL invite applications for Free Online Course on Deep Learning

IIT Ropar has organised this course as deep Learning has received a lot of attention over the past few years and has been employed successfully by companies like Google, Microsoft, IBM, Facebook, Twitter etc. It helps them to solve a wide range of problems in Computer Vision and Natural Language Processing.

In this course, IIT Ropar prof will teach about the building blocks used in these Deep Learning solutions specifically, about feedforward neural networks, convolutional neural networks, recurrent neural networks and attention mechanisms. Students will also look at various optimization algorithms such as Gradient Descent, Nesterov Accelerated Gradient Descent, Adam, AdaGrad and RMSProp which are used for training such deep neural networks. At the end of this course, students would, thus, have knowledge of deep architectures used for solving various Vision and NLP tasks.

Who can enrol in the course?

This is undergraduate computer science and engineering level course. But anyone can enrol in the course. The only prerequisite is you should have a working knowledge of Linear Algebra and Probability Theory. Further, it would be beneficial if the participants have done a course on Machine Learning. You also should have the desire to expand the horizon of your knowledge.

Timeline of the course

This Elective course has a duration of 12 weeks. It will start from 18 January 2021 and end on 09 April 2021. But, if you want the certificate, you have to give a proctored exam on 25 April 2021. The last date to enrol is 25 January 2020.

Who will teach this Deep Learning course?

Prof. Sudarshan will lead the course. Sudarshan Iyengar has a PhD from the Indian Institute of Science. Currently, he is working as an assistant professor at IIT Ropar. Also, he has been teaching this course from the past 4 years.

Edit as suggested by a faculty from IIT Madras: Prof Mitesh Khapra of IIT Madras originally created the content. So, the videos are from him. Prof Sudarshan is coordinating and offering it. Thus, the course was originally offered by Prof Mitesh Khapra and is now being offered by Prof Sudarshan.

What will the course teach?

This is a 12-week-long course. Thus, here’s the itinerary for the course.

W1: (Partial) History of Deep Learning, Deep Learning Success Stories, McCulloch Pitts Neuron, Thresholding Logic,

Perceptrons, Perceptron Learning Algorithm

W2: Multilayer Perceptrons (MLPs), Representation Power of MLPs, Sigmoid Neurons, Gradient Descent, Feedforward

Neural Networks, Representation Power of Feedforward Neural Networks

W3: FeedForward Neural Networks, Backpropagation

W4: Gradient Descent (GD), Momentum Based GD, Nesterov Accelerated GD, Stochastic GD, AdaGrad, RMSProp, Adam,

Eigenvalues and eigenvectors, Eigenvalue Decomposition, Basis

W5: Principal Component Analysis and its interpretations, Singular Value Decomposition

W6: Autoencoders and relation to PCA, Regularization in autoencoders, Denoising autoencoders, Sparse autoencoders,

Contractive autoencoders

W7: Regularization: Bias Variance Tradeoff, L2 regularization, Early stopping, Dataset augmentation, Parameter sharing

and tying, Injecting noise at the input, Ensemble methods, Dropout

W8: Greedy Layerwise Pre-training, Better activation functions, Better weight initialization methods, Batch Normalization

W9: Learning Vectorial Representations Of Words

W10: Convolutional Neural Networks, LeNet, AlexNet, ZF-Net, VGGNet, GoogLeNet, ResNet, Visualizing Convolutional

Neural Networks, Guided Backpropagation, Deep Dream, Deep Art, Fooling Convolutional Neural Networks

W11: Recurrent Neural Networks, Backpropagation through time (BPTT), Vanishing and Exploding Gradients, Truncated BPTT, GRU, LSTMs

W12: Encoder-Decoder Models, also, Attention Mechanism, Attention over images

How to obtain a certificate from IIT Ropar?

The course is free to enrol and learn. But if you want a certificate, you have to register and write the optional proctored exam. The fee for this exam is ₹ 1000. Also, the successful completion of the exam does not guarantee a certificate. Further, to get a certificate, you need to get 25% from the assignments and 75% of the proctored certification exam score out of 100.

Final score = Average assignment score (>10/25) + Exam score (>30/75). If one of the 2 criteria is not met, you will not get the certificate even if the Final score > 40/100.

This printable certificate will carry the stamp from both NPTEL and IIT Ropar.

Further, enrol in the course here.