COURSE OVERVIEW
Introduction to Deep Learning
• What is Deep Learning?
• AI vs ML vs DL
• Applications of Deep Learning
• History and Evolution of Neural Networks
• Neural Networks vs Traditional ML
Math Foundation for Deep Learning
• Linear Algebra
• Vectors, Matrices, Tensors
• Matrix Multiplication
• Calculus
• Derivatives & Gradients
• Chain Rule (backpropagation concept)
• Probability and Statistics
• Probability Distributions
• Bayes Theorem
• Optimization Concepts
• Gradient Descent (Batch, Mini-Batch, Stochastic)
• Cost Functions
• Learning Rate and Convergence
Python & Deep Learning Libraries Setup
• Setting up Conda/Jupyter for DL
• Installing TensorFlow, PyTorch, Keras
• Working with NumPy
• Working with Pandas
• Visualization using Matplotlib and Seaborn
• Google Colab vs Local GPU vs Cloud GPU
Artificial Neural Networks (ANN)
• Perceptron & Multi-Layer Perceptron (MLP)
• Activation Functions: Sigmoid, ReLU, Tanh, Softmax
• Feedforward and Backpropagation
• Loss Functions: MSE, Cross-Entropy
• Epochs, Batches, Iterations
• Model Training Pipeline
• Overfitting and Underfitting
• Regularization: L1, L2, Dropout
• Model Evaluation Metrics (Accuracy, Precision, Recall, F1, etc.)
Convolutional Neural Networks (CNNs)
• CNN Architecture
• Convolution Layers, Filters, Feature Maps
• Pooling Layers: MaxPooling, AvgPooling
• Flattening and Fully Connected Layers
• Padding and Strides
• Image Preprocessing Techniques
• Transfer Learning
• ResNet
• VGG
• Inception
• CNN Projects
• Image Classification
• Object Detection (basic intro)
• Face Recognition
Recurrent Neural Networks (RNNs)
• Why Sequence Models?
• RNN Architecture & Working
• Vanishing Gradient Problem
• Long Short-Term Memory (LSTM)
• Gated Recurrent Units (GRU)
• Applications of RNN
• Text Classification
• Time Series Forecasting
• Sentiment Analysis
Course Outcomes
-
Understand and apply deep learning concepts including neural networks, CNNs, RNNs, and LSTMs.
-
Develop and train models using frameworks like TensorFlow and PyTorch for real-world tasks.
-
Implement applications in computer vision, NLP, and time-series analysis for industry use.
-
Optimize and evaluate models for performance, accuracy, and scalability in production environments.
-
Gain hands-on experience to build job-ready deep learning projects and solutions.
