Open Source Knowledge Platform

AI Algo Hub

Master every AI algorithm from linear regression to diffusion models. 80+ algorithm implementations from scratch, interactive curriculum, and a decision framework to pick the right algorithm for your problem.

80+Algorithms
6Categories
40+Code Examples
17Tracks

Explore by Category

From classic ML to cutting-edge generative AI

🤖

Classic ML Algorithms

Linear Regression, Logistic Regression, Decision Trees, SVM, KNN, Naive Bayes, Random Forest, Gradient Boosting, K-Means, PCA, t-SNE, UMAP, and more.

22+ Algorithms
🧠

Deep Learning

Neural Networks, CNNs, RNNs/LSTM/GRU, Autoencoders, GANs, Diffusion Models, Graph Neural Networks, and advanced architectures.

15+ Architectures
🚀

LLM & Modern AI

Attention, Fine-tuning, LoRA/PEFT, RAG, Tokenization (BPE), Prompt Engineering, Quantization, Contrastive Learning, MLOps.

14+ Topics
🎯

Algorithm Picker

Don't know which algorithm to use? Answer a few questions about your data and problem, and get the right algorithm recommendation.

Decision Framework
📚

Learning Curriculum

Structured 80-module learning path from statistics fundamentals to diffusion models, with pure Python implementations from scratch.

80 Modules
📱

Android Development

Jetpack Compose, Material Design 3, MVVM, Coroutines, Hilt DI, Room, Retrofit, Performance Optimization, and Testing.

25+ Articles

⚡ Quick Algorithm Decision Guide

1 What's your goal? Predict a value (Regression), Classify (Classification), Group data (Clustering), Reduce dimensions, Generate content, or Make decisions (RL)?
2 How much data do you have? Small (<1K): Simple models. Medium (1K-100K): Tree-based. Large (100K+): Deep learning.
3 Do you need interpretability? Yes: Linear models, Decision Trees, SHAP. No: Ensemble methods, Neural Networks.
4 What type of data? Tabular: XGBoost/LightGBM. Images: CNN/ViT. Text: Transformers. Sequences: LSTM/Transformer. Graph: GNN.

Featured Algorithms

Supervised

Linear Regression

Predict continuous values by fitting a linear relationship between features and target. The foundation of predictive modeling.

Regression Beginner Interpretable
Supervised

XGBoost / Gradient Boosting

The king of tabular data. Sequentially builds weak learners that correct previous errors. Wins most Kaggle competitions.

Classification Regression Advanced
Deep Learning

Transformers

Self-attention based architecture that revolutionized NLP and now dominates vision, audio, and multimodal tasks.

NLP Vision Expert
Deep Learning

CNN (Convolutional Neural Networks)

Specialized for spatial data using learnable filters. Powers image classification, object detection, and segmentation.

Computer Vision Intermediate
Supervised

Random Forest

Ensemble of decision trees using bagging. Robust, handles missing data, and requires minimal tuning.

Classification Regression Intermediate
Generative

Diffusion Models (DDPM)

Generate high-quality images by learning to reverse a noise process. Behind Stable Diffusion, DALL-E, and Midjourney.

Image Generation Expert
Unsupervised

K-Means Clustering

Partition data into K clusters by minimizing within-cluster variance. Simple, fast, and widely used for customer segmentation.

Clustering Beginner
NLP / LLM

RAG (Retrieval-Augmented Generation)

Combine LLMs with external knowledge retrieval for grounded, up-to-date answers without retraining.

LLM Production Advanced

Learning Path Preview

80 modules from foundations to frontier algorithms

Part I
Statistics & Data Foundations
Beginner
Part II
Evaluation Metrics
Beginner
Part III
Linear Models
Intermediate
Part IV
Classic ML Algorithms
Intermediate
Part V
Neural Networks & Advanced
Intermediate
Part VIII
Deep Learning Track
Advanced
Part XI
Frontier Algorithms
Expert
Part XVII
Generative Systems
Expert
AGENT_CONTEXT: AI Algo Hub - Comprehensive AI/ML algorithm knowledge platform. CATEGORIES: ml-algorithms, deep-learning, llm-modern-ai, algorithm-picker, curriculum, android-dev. ALGORITHMS: 80+ algorithms from linear regression to diffusion models. CURRICULUM: 80-module learning path with pure Python implementations. API: /api/search, /api/algorithm/:id, /api/recommend