Udemy Deep Learning A-Z Course

General Concepts

• Understanding of Artificial Intelligence, Machine Learning, and Deep Learning
• Basics of Neural Networks
• Activation Functions (e.g., ReLU, Sigmoid, Tanh)
• Loss Functions (e.g., Mean Squared Error, Cross-Entropy)
• Optimization Algorithms (e.g., Gradient Descent, Adam)
• Overfitting and Regularization Techniques (e.g., Dropout, L1/L2 Regularization)
• Data Preprocessing and Augmentation
• Understanding of Hyperparameters and Their Tuning
• Model Evaluation Metrics (e.g., Accuracy, Precision, Recall)

Artificial Neural Networks (ANNs)

Developed

1. an Artificial Neural Network to predict the net hourly electrical output based on features like temperature, humidity, pressure, the exhaust vacuum, etc.
2. an Artificial Neural Network to predict if a particular bank customer will churn from the bank.

Concepts Learned:

• Structure of ANNs (Neurons, Layers, Weights, Biases)
• Forward Propagation and Backpropagation
• Batch Normalization
• Weight Initialization Techniques

Convolutional Neural Networks (CNNs)

Developed a Dog vs Cat CNN Classifier. Built my own streamlit app (Demo video available here here)

Concepts Learned:

• Understanding Convolutional Layers
• Pooling Layers (Max Pooling, Average Pooling)
• Stride and Padding Concepts
• Applications of CNNs (e.g., Image Recognition, Object Detection)

Recurrent Neural Networks (RNNs)

Developed an LSTM (Long Short-Term Memory) Recurrent Neural Network to predict stock prices

Concepts Learned:

• Structure and Functioning of RNNs
• Problems of Vanishing and Exploding Gradients
• Long Short-Term Memory (LSTM) Networks
• Applications of RNNs (e.g., Time Series Prediction, Natural Language Processing)