NeuralPath LogoNeuralPath

Deep Learning

Understand multi-layer perceptrons, activation functions, backpropagation, CNNs, and modern training loops.

Your Progress0 / 5 completed
0%
Continue Learning

Foundations

(1 topic)

Perceptrons and Multilayer Neural Networks

From a single artificial neuron to deep Multilayer Perceptrons (MLPs)—discover how stacked layers, hidden representations, and PyTorch nn.Module solve complex non-linear problems.

Neural Networks

(1 topic)

Activation Functions

Why neural networks need non-linear switches—from Sigmoid and Tanh to ReLU, Leaky ReLU, Softmax, and modern LLM activations like GELU and SwiGLU.

Training & Optimization

(2 topics)

Backpropagation

How neural networks learn from their mistakes—from intuitive blame assignment and the Chain Rule to matrix backpropagation and advanced memory optimization.

Optimizers and Learning-Rate Schedules

How neural networks turn raw gradients into smart weight updates—from SGD and Momentum to AdamW, Warmup, and Cosine Annealing schedules.

Computer Vision

(1 topic)

CNNs and Image Recognition

How Convolutional Neural Networks see the world—from sliding filters, stride, and pooling to ResNet skip connections and modern Computer Vision.