Neural Networks and Deep Learning

77 lessons in AI Fundamentals

• Selecting the LSTM Last TimestepCode Quiz
• MaxPool Stride and DownsamplingCode Quiz
• Weight Scale and Exploding ValuesCode Quiz
• Data Leakage in ScalingCode Quiz
• Learning Rate Scheduler PlacementCode Quiz
• Batch Normalization FormulaCode Quiz
• Weight Initialization StrategyCode Quiz
• Cross-Entropy Loss ComputationCode Quiz
• Dropout Behavior During EvaluationCode Quiz
• PyTorch Training LoopCode Quiz
• Forward Pass Through a LayerCode Quiz
• SGD Parameter UpdateCode Quiz
• Gradient of a Squared ErrorCode Quiz
• Sigmoid Activation ImplementationCode Quiz
• Configuring the Adam OptimizerCode Quiz
• Feedforward Network Output LayerCode Quiz
• Counting Conv Layer ParametersCode Quiz
• Conv Layer Input ChannelsCode Quiz
• Counting Iterations per EpochCode Quiz
• Normalizing Test Data CorrectlyCode Quiz
• Weights, Biases, and Weighted SumsSlides / Video
• The Artificial Neuron / PerceptronSlides / Video
• What a Neural Network IsSlides / Video
• Real-World Applications of Deep LearningSlides / Video
• CNN vs RNN Architectures OverviewSlides / Video
• Regularization: Dropout and Weight DecaySlides / Video
• Feature Learning vs Manual Feature EngineeringSlides / Video
• What Makes a Network 'Deep'Slides / Video
• Epochs, Batches, and IterationsSlides / Video
• Hierarchical Feature LearningQuiz
• Purpose of Activation FunctionsQuiz
• Epochs, Batches, IterationsQuiz
• CNNs OverviewQuiz
• Learning Rate EffectQuiz
• Why Non-Linearity MattersQuiz
• Backpropagation ConceptQuiz
• Weight Initialization ImportanceQuiz
• Perceptron StructureQuiz
• Gradient Descent TrainingQuiz
• Vanishing GradientsQuiz
• Forward PropagationQuiz
• Weights and BiasesQuiz
• Biological Neuron InspirationQuiz
• Network LayersQuiz
• What Makes a Network DeepQuiz
• Weighted Sum ComputationQuiz
• Common Activation FunctionsQuiz
• Fully Connected LayersQuiz
• RNNs OverviewQuiz
• Dropout RegularizationQuiz
• Learning Rate and Its Effect on TrainingSlides / Video
• Backpropagation BasicsSlides / Video
• Gradient Descent for OptimizationSlides / Video
• Loss Functions for Neural NetworksSlides / Video
• Forward Propagation BasicsSlides / Video
• Neural Network Layers ExplainedSlides / Video
• Why Deep Learning Needs Data and ComputeFlashcard
• Recurrent Neural Networks (RNNs)Flashcard
• Convolutional Neural Networks (CNNs)Flashcard
• Artificial Neuron / Perceptron StructureFlashcard
• Loss / Cost FunctionsFlashcard
• Activation Functions: PurposeFlashcard
• Weights, Biases, and Weighted SumFlashcard
• Vanishing and Exploding GradientsFlashcard
• Epochs, Batches, and IterationsFlashcard
• Input, Hidden, and Output LayersFlashcard
• Gradient Descent IntuitionFlashcard
• BackpropagationFlashcard
• What Makes a Network 'Deep'Flashcard
• Learning RateFlashcard
• Hyperparameters vs Learned ParametersFlashcard
• Common Activation FunctionsFlashcard
• Feature / Representation LearningFlashcard
• Regularization (Dropout, etc.)Flashcard
• Neural Networks and Biological InspirationFlashcard
• Feedforward and Forward PropagationFlashcard
• Activation Functions: Sigmoid, ReLU, TanhSlides / Video