DEEP LEARNING PROGRAM

Overview:

With the Deep Learning Program, you will master the fundamentals of deep learning and be better equipped to contribute to the advancement of cutting-edge AI technologies.

Building and training neural network designs including Convolution Neural Networks, Recurrent Neural Networks, LSTM/Transformers and more will be covered in this Specialization. You will learn how to improve them using tactics like Dropout, Batch Norm and more. As a Python and TensorFlow expert, you’ll learn about theoretical principles and their practical applications in a variety of industries. You’ll also learn how to solve real-world problems like speech recognition and music synthesis using these tools.

In this course, you will learn how to build neural networks with PyTorch, a deep learning framework for Python. Incorporate convolutional networks, recurrent networks, generative adversarial networks, and web-deployable models for image recognition, sequence generation, and picture production, respectively.

Many industries are being transformed by AI. The Deep Learning Specialization enables you to take your profession to the next level by helping you acquire the necessary knowledge and abilities. You’ll also get career guidance from specialists in deep learning from both business and academia as you go.

Key Learnings:

  • Understand Keras and TensorFlow ideas, as well as their core functions, operations, and execution pipeline.
  • Implement deep learning algorithms, learn how to use neural networks, and navigate the many layers of data abstraction.
  • Advanced subjects such as convolutional neural networks, recurrent neural networks, training deep networks, and high-level interfaces must be mastered and comprehended.
  • Create and understand deep learning models using the Keras and TensorFlow frameworks.
  • Understand the vocabulary and basic principles of artificial neural networks, as well as the use of autoencoders and Pytorch.
  • Deep learning models should be troubleshooted and improved.
  • Create a deep learning project of your own.
  • Distinguish between the terms “machine learning,” “deep learning,” and “artificial intelligence.”

Course Modules

  • Introduction
  • What is AI and Deep Learning
  • Brief History of AI
  • Recap: SL, UL and RL
  • Deep Learning: Successes Last Decade
  • Demo and Discussion: Self-Driving Car Object Detection
  • Applications of Deep Learning
  • Challenges of Deep Learning
  • Demo and Discussion: Sentiment Analysis Using LSTM
  • Full Cycle of a Deep Learning Project
  • Key Takeaways
  • Knowledge Check
  • Biological Neuron Vs Perceptron
  • Shallow Neural Network
  • Training a Perceptron
  • Demo Code #1: Perceptron (Linear Classification)
  • Backpropagation
  • Role of Activation Functions and Backpropagation
  • Demo Code #2: Activation Function
  • Demo Code #3: Backprop Illustration
  • Optimization
  • Regularization
  • Dropout layer
  • Demo Code #4: Dropout Illustration, Lesson-end Exercise (Classification Kaggle Dataset)
  • Key Takeaways
  • Knowledge Check
  • Lesson-end Project
  • Deep Neural Network: Why and Applications
  • Designing a Deep Neural Network
  • How to Choose Your Loss Function?
  • Tools for Deep Learning Models
  • Keras and its Elements
  • Demo Code #5: Build a Deep Learning Model Using Keras
  • Tensorflow and Its Ecosystem
  • Demo Code #6: Build a Deep Learning Model Using Tensorflow
  • TFlearn
  • Pytorch and its Elements
  • Demo Code #7: Build a Deep Learning Model Using Pytorch
  • Demo Code #8: Lesson-end Exercise
  • Key Takeaways
  • Knowledge Check
  • Lesson-end Project
  • Optimization Algorithms
  • SGD, Momentum, NAG, Adagrad, Adadelta , RMSprop, Adam
  • Demo code #9: MNIST Dataset
  • Batch Normalization
  • Demo Code #10
  • Exploding and Vanishing Gradients
  • Hyperparameter Tuning
  • Demo Code #11
  • Interpretability
  • Demo Code#12: MNIST– Lesson-end Project with Interpretability Lessons
  • Width vs Depth
  • Key Takeaways
  • Knowledge Check
  • Lesson-end Project
  • Success and History
  • CNN Network Design and Architecture
  • Demo Code #13: Keras
  • Demo Code #14: Two Image Type Classification (Kaggle), Using Keras
  • Deep Convolutional Models
  • Key Takeaways
  • Knowledge Check
  • Lesson-end Project
  • Sequence Data
  • Sense of Time
  • RNN Introduction
  • Demo Code #15: Share Price Prediction with RNN
  • LSTM (Retail Sales Dataset Kaggle)
  • Demo Code #16:
  • Word Embedding and LSTM
  • Demo Code #17: Sentiment Analysis (Movie Review)
  • GRUs
  • LSTM vs GRUs
  • Demo Code #18: Movie Review (Kaggle), Lesson-end Project)
  • Key Takeaways
  • Knowledge Check
  • Lesson-end Project
  • Introduction to Autoencoders
  • Applications of Autoencoders
  • Autoencoder for Anomaly Detection
  • Demo Code #19: Autoencoder Model for MNIST Data
  • Key Takeaways
  • Knowledge Check
  • Lesson-end Project