Artificial Intelligence with Deep Learning Workshop

Overview

Learn How to Develop Neural Networks From Scratch and Develop and Deploy Models.

There is a technological revolution happening, changing all aspects of our daily lives. AI (Artificial Intelligence) has penetrated deep into our activities, interactions, professions, comforts, and experiences.

Deep learning is a machine learning technique that teaches computers to do what comes naturally to humans: learn by example. Deep learning is a key technology behind driverless cars, enabling them to recognize a stop sign or to distinguish a pedestrian from a lamppost. It is the key to voice control in consumer devices like phones, tablets, TVs, and hands-free speakers. Deep learning is getting lots of attention lately and for good reason. It’s achieving results that were not possible before.

In deep learning, a computer model learns to perform classification tasks directly from images, text, or sound. Deep learning models can achieve state-of-the-art accuracy, sometimes exceeding human-level performance. Models are trained by using a large set of labeled data and neural network architectures that contain many layers. This immersive deep learning course teaches you how to use this data to create outcomes to use in your development.

In This Artificial Intelligence and Deep Leaning Course You Will: 

• Understand the basics of deep learning
• Use Tuning Models
• Create convolutional neural networks
• Use Recurrent neural networks
• Develop and deploy models
• Use python / Jupyter notebooks
• Know the best practices to follow

Course Outline

Part 1: Introduction to Deep Learning

1. What is a neural network?
2. Supervised Learning with Neural Networks
3. Why is Deep Learning taking off?

Part 2: Neural Networks Basics

1. Binary Classification
2. Logistic Regression
3. Logistic Regression Cost Function
4. Gradient Descent
5. Derivatives
6. More Derivative Examples
7. Computation graph
8. Derivatives with a Computation Graph
9. Logistic Regression Gradient Descent
10. Gradient Descent on m Examples
11. Vectorization

Part 3: Shallow Neural Networks

1. Neural Networks Overview
2. Neural Network Representation
3. Computing a Neural Network’s Output
4. Vectorizing across multiple examples
5. Explanation for Vectorized Implementation
6. Activation functions
7. Why do you need non-linear activation functions?
8. Derivatives of activation functions
9. Gradient descent for Neural Networks

Part 4: Deep Neural Networks

1. Deep L-layer neural network
2. Forward Propagation in a Deep Network
3. Getting your matrix dimensions right
4. Why deep representations?
5. Building blocks of deep neural networks
6. Forward and Backward Propagation
7. Parameters vs Hyperparameters

Part 5: Practical Aspects of Deep Learning

1. Train / Dev / Test sets
2. Bias / Variance
3. Basic Recipe for Machine Learning
4. Regularization
5. Why regularization reduces overfitting?
6. Dropout Regularization
7. Understanding Dropout
8. Other regularization methods
9. Normalizing inputs
10. Vanishing / Exploding gradients
11. Weight Initialization for Deep Networks
12. Numerical approximation of gradients
13. Gradient checking
14. Gradient Checking Implementation Notes

Part 6: Optimization Algorithms

1. Mini-batch gradient descent
2. Understanding mini-batch gradient descent
3. Exponentially weighted averages
4. Understanding exponentially weighted averages
5. Bias correction in exponentially weighted averages
6. Gradient descent with momentum
7. RMSprop
8. Adam optimization algorithm
9. Learning rate decay
10. The problem of local optima

Part 7: Hyperparameter Tuning, Batch Normalization, and Programming Frameworks

1. Tuning process
2. Using an appropriate scale to pick hyperparameters
3. Hyperparameters tuning in practice: Pandas vs. Caviar
4. Normalizing activations in a network
5. Fitting Batch Norm into a neural network
6. Why does Batch Norm work?
7. Batch Norm at test time
8. Softmax Regression
9. Training a softmax classifier
10. Deep learning frameworks
11. TensorFlow

Part 8: Foundations of Convolutional Neural Networks

1. Computer Vision
2. Edge Detection Example
3. More Edge Detection
4. Padding
5. Strided Convolutions
6. Convolutions Over Volume
7. One Layer of a Convolutional Network
8. Simple Convolutional Network Example
9. Pooling Layers
10. CNN Example
11. Why Convolutions?
12. Deep convolutional models: case studies
13. Classic Networks
14. ResNets
15. Why ResNets Work
16. Networks in Networks and 1×1 Convolutions
17. Inception Network Motivation
18. Inception Network
19. Using Open-Source Implementation
20. Transfer Learning
21. Data Augmentation
22. State of Computer Vision

Part 9: Recurrent Neural Networks

1. Why sequence models
2. Notation
3. Recurrent Neural Network Model
4. Backpropagation through time
5. Different types of RNNs
6. Language model and sequence generation
7. Sampling novel sequences
8. Vanishing gradients with RNNs
9. Gated Recurrent Unit (GRU)
10. Long Short Term Memory (LSTM)
11. Bidirectional RNN
12. Deep RNNs

Part 10: Natural Language Processing & Word Embeddings

1. Word Representation
2. Using word embeddings
3. Properties of word embeddings
4. Embedding matrix
5. Learning word embeddings
6. Word2Vec
7. Negative Sampling
8. GloVe word vectors
9. Sentiment Classification
10. Debiasing word embeddings
11. Sequence models & Attention mechanism
12. Basic Models
13. Picking the most likely sentence
14. Beam Search
15. Refinements to Beam Search
16. Error analysis in beam search
17. Bleu Score (optional)
18. Attention Model Intuition
19. Attention Model
20. Speech recognition
21. Trigger Word Detection

Part 11: ML Strategy

1. Why ML Strategy
2. Orthogonalization
3. Single number evaluation metric
4. Satisficing and Optimizing metric
5. Train/dev/test distributions
6. Size of the dev and test sets
7. When to change dev/test sets and metrics
8. Why human-level performance?
9. Avoidable bias
10. Understanding human-level performance
11. Surpassing human-level performance
12. Improving your model performance
13. Carrying out error analysis
14. Cleaning up incorrectly labeled data
15. Build your first system quickly, then iterate
16. Training and testing on different distributions
17. Bias and Variance with mismatched data distributions
18. Addressing data mismatch
19. Transfer learning
20. Multi-task learning
21. What is end-to-end deep learning?
22. Whether to use end-to-end deep learning