Deep Learning. a subset of machine learning, encompasses neural networks that can learn from raw or unstructured data, much like humans. It’s used for speech recognition, machine translation, computer vision and natural language processing. Deep Learning has applications in medical diagnosis, server optimisation, data centre security, autonomous driving and more. 

Below, we have listed down seven resources to learn Deep Learning.

Continuous learning at Association of Data Scientists

The Association of Data Scientists offers online courses to provide in-depth knowledge of various areas within machine learning and data science. Most of these courses are available as videos for self-paced learning along with relevant Colab notebooks. 

ADaSci offers a Data Science for Beginners tutorial, explaining different landscapes in artificial intelligence, basics of Python programming, and the importance of data analyses and visualisations. It also includes a hands-on session on regression, classification and clustering. 

ADaSci also offers a Comprehensive Guide to 15 Most Important NLP Datasets (e-book), and workshops on Meta Learning, Computer Vision and Deployment and Model Management.

For more information, click here. 

Deep Learning Specialisation: Coursera 

The course offered by DeepLearning.AI is designed by Andrew Ng, Kian Katanforoosh, and Younes Bensouda Mourri. It is a foundational programme for intermediaries to master the fundamentals of deep learning and understand the capabilities and challenges. 

As the name suggests, the Deep Learning Specialisation course, as the name suggests, assists learners to train test sets, analyse variance for DL applications, use standard techniques and optimisation algorithms, and build neural networks in TensorFlow. 

You will get to learn how to build and train deep neural network architectures such as Convolutional Neural Networks, Recurrent Neural Networks, LSTMs, Transformers, and how to make them better with strategies like Dropout, BatchNorm, and Xavier/He initialisation. Additionally, the course will help learners master theoretical concepts and their industry applications using Python and TensorFlow. 

The attendees must have a basic understanding of linear algebra and ML and intermediate-level Python skills.

For more information, click here. 

Deep Learning: NYC 

The New York University Center for Data Science offers a course on Deep Learning, delivered by Yann LeCun and Alfredo Canziani. It teaches students the latest techniques in deep learning and representation learning. The course focuses on supervised and unsupervised deep learning and covers metric learning, convolutional and recurrent nets, computer vision applications, natural language understanding, and speech recognition. 

The course offers sessions on the history and evolution of deep learning, parameter transformation, natural signals’ properties, optimisation, training RNNs, Predictive Control, structured prediction and regularisation. 

The ideal candidate should have completed a graduate-level machine learning course. 

For more information, click here. 

The Complete Deep Learning Course: Udemy 

The Complete Deep Learning Course 2021, created by Hoang Quy La, is meant for anyone interested in deep learning. The course teaches you to use TensorFlow with Python. You will learn about Artificial Neural Networks, Recurrent Neural Networks, Deep Convolutional Generative adversarial network, image processing, autoencoder, Monte Carlo method of Deep Reinforcement Learning, Convolution Neural Network, Generative Adversarial Network, Natural Language Processing, Sentiment Analysis and Restricted Boltzmann Machine. 

Additionally, the course will provide hands-on experience on concrete quality prediction using deep neural networks, CIFAR-10, classifying clothing models, predicting stock price and Iris flower. 

While a basic understanding of Python will come in handy, the course does not come with any prerequisites. High school knowledge in mathematics is enough to take up this course. 

For more information, click here. 

Introduction to Deep Learning: MIT 

MIT’s introductory course on Deep Learning will teach applications of Computer Vision and Natural Language Processing. Alexander Amini and Ava Soleimany lead the course. You will gain foundational knowledge of deep learning algorithms and get hands-on experience building neural networks in TensorFlow. 

The course covers TensorFlow and building NNs from scratch, music generation using RNNs, image classification and detection, evidential deep learning, debiasing facial recognition systems, and bias and fairness. Students will be able to participate in a project proposal competition with feedback from staff and a panel of industry experts at the end of the course. 

Attendees should have a basic understanding of calculus and linear algebra. While a prior experience is desirable, it is not necessary. 

For more information, click here. 

Deep Learning Nanodegree program: Udacity 

The Nanodegree program on Deep Learning is meant for beginners who want to implement neural networks using PyTorch. At the end of the course, you will be able to build convolutional networks for image recognition, recurrent networks for sequence generation, deploy sentimental analysis models, generative adversarial networks for image generation and sequence generation. 

The beginner-friendly programme is meant for students with working knowledge of Python, including NumPy and pandas. Familiarity with calculus and linear algebra is preferable. 

For more information, click here. 

Practical Deep Learning for coders: Fast.ai

Sylvain Gugger and Jeremy Howard, developers of Fast.ai, have set up a website Practical Deep Learning for Coders.

The course covers computer vision (image classification), image localisation and detection, natural language processing, and collaborative filtering. You will also learn how to turn the models into web applications, how deep learning models work, and the latest deep learning techniques. The course uses PyTorch and fast.ai. 

The attendees should know the basics of coding, preferably in Python. 

For more information, click here. 

An Introduction to Practical Deep Learning: Coursera 

The introductory course to Deep Learning is offered by Intel and is delivered by Andres Rodriguez, Nikhil Murthy, and Hanlin Tang. 

The course covers deep learning basics, Convolutional Neural Networks, Fine Tuning and detection, Recurrent Neural Networks, Training tips and Multimode Distributed Training. The course will explore deep learning concepts, focusing on training deep networks using Intel Nervana Neon.

For more information, click here.