3 learning groups / 18 visual topics

Learn deep learning.
Understand the system behind it.

A refined visual learning space for deep learning concepts - calm, structured, and easy to explore, with a clear path from fundamentals to modern architectures.

Start for free Browse all topics

18Visual topics

3Core groups

8Visual notes

0Prerequisites

Course groups

The most important deep learning topics, arranged as one elegant learning path.

Start with core training intuition, move through vision and sequence models, then sharpen practical diagnosis skills.

Module 01

Deep Learning Foundations & Training

From a single neuron to backpropagation, optimization, regularization, and training behavior.

Single Neuron Forward Pass→Activation Functions Comparison→Loss Functions→Backpropagation→Gradient Descent & Learning Rate→Overfitting vs. Underfitting→Dropout→Adam Optimizer vs. SGD→View all 8 topics →

Module 02

Vision & Sequence Models

Connect CNN visual intuition with sequence models and attention.

Convolution Operation→Pooling and Downsampling→Feature Map Visualization→RNN Structure→Attention Mechanism Intuition→View all 5 topics →

Module 03

Model Development Toolkit

Diagnose data splits, metrics, bias, variance, mini-batches, and transfer decisions.

Train / Val / Test Split→Evaluation Metrics & Confusion Matrix→Bias vs. Variance Diagnosis→Mini-batch Training & Batch Size→Transfer Learning Intuition→View all 5 topics →

Why it works

Designed with one clear intention.

Simple at first glance.

Fewer, stronger visual blocks guide your eye naturally without dashboard clutter.

Elegant typography.

Large headlines, quieter body text, and deliberate spacing create a clearer learning identity.

Deep learning focused.

Every visual is tied to a real topic, so style supports the substance behind it.

Learn deep learning.Understand the system behind it.