Interactive guide
AI Algorithms and Models Guide
Clear explanations and interactive demonstrations of the algorithms, models and methods used in artificial intelligence and machine learning.
Explore by family
Start from a broad area, then open the methods in it. Some methods appear in more than one family.
Prediction and classification
Methods that learn from labeled examples to predict a number or assign a case to a group.
- Linear Regression
- Logistic Regression
- Decision Tree
- Random Forest
- Gradient Boosting
- KNN (Neighbors)
- Naive Bayes
- SVM (The Divider)
Regression
Linear regression predicts a number; logistic regression estimates the chance of a yes-or-no outcome.
Patterns, anomalies and recommendations
Methods that find structure in data without a fixed answer: groups, unusual cases, items that go together and recommendations.
Language, images and audio
Language and text models
Methods that represent words, read text in context, and generate or transform language.
Computer vision
Methods that find, outline, read and match what is in an image.
Audio and speech
Methods that turn speech into text and learn from raw audio.
Image and content generation
Models that create new text or images, change an image's style, and the risks that come with generated media.
Search, planning and decision-making
Search and retrieval
Methods that score and rank documents for a query, and systems that look up sources before they answer.
Reinforcement learning, agents and planning
Methods that learn from rewards, search ahead before acting, plan routes and work through tasks step by step.
Neural networks and deep learning
The building blocks behind most modern models: layered networks, memory for sequences and networks for connected data.
Forecasting and optimization
Methods that predict values over time, estimate how long until an event, and find the best plan within limits.
Showing all 58 entries
1 / Prediction and classification
Decision Tree
A decision tree is like a question game. It asks simple yes-or-no questions until it reaches an answer.
Open try-it page →2 / Prediction and classification
Random Forest
A random forest uses many decision trees. Each tree votes, and the answer with the most votes wins.
Open try-it page →3 / Prediction and classification
Linear Regression
Linear regression draws a straight line through dots so it can guess a number.
Open try-it page →4 / Prediction and classification
Logistic Regression
Logistic regression turns clues into a chance, then chooses yes or no.
Open try-it page →5 / Prediction and classification
KNN (Neighbors)
KNN looks at the closest examples and copies the group most of them belong to.
Open try-it page →6 / Prediction and classification
Naive Bayes
Naive Bayes uses word clues and math chance to sort text quickly.
Open try-it page →7 / Prediction and classification
Gradient Boosting
Gradient boosting builds one small model after another. Each new model tries to fix the last mistakes.
Open try-it page →8 / Prediction and classification
SVM (The Divider)
SVM draws a line between two groups and tries to leave the biggest empty space between them.
Open try-it page →9 / Patterns, anomalies and recommendations
K-Means
K-Means groups dots by putting them near the closest center point.
Open try-it page →10 / Patterns, anomalies and recommendations
DBSCAN
DBSCAN finds groups where many dots are close together and marks lonely dots as noise.
Open try-it page →11 / Patterns, anomalies and recommendations
Isolation Forest
Isolation Forest finds strange items by seeing which ones are easy to separate from the rest.
Open try-it page →12 / Patterns, anomalies and recommendations
Collaborative Filtering
Collaborative filtering recommends things by looking at people with tastes like yours.
Open try-it page →13 / Patterns, anomalies and recommendations
PCA (Data Shrinker)
PCA makes a huge table smaller while trying to keep the most important pattern.
Open try-it page →14 / Patterns, anomalies and recommendations
Apriori Algorithm
Apriori finds items that often appear together.
Open try-it page →15 / Computer vision
YOLO (You Only Look Once)
YOLO looks at one picture and quickly finds objects in it.
Open try-it page →16 / Computer vision
Faster R-CNN
Faster R-CNN first finds places where objects might be, then checks those places carefully.
Open try-it page →17 / Computer vision
Mask R-CNN
Mask R-CNN finds objects and colors the exact pixels that belong to each object.
Open try-it page →18 / Computer vision
ResNet
ResNet helps a very deep AI pass information forward without getting lost.
Open try-it page →19 / Computer vision
Vision Transformer (ViT)
A Vision Transformer cuts a picture into small squares and checks how the squares relate to each other.
Open try-it page →20 / Computer vision
Pose Estimation
Pose estimation finds body points like elbows and knees so it can understand how a person is standing or moving.
Open try-it page →21 / Computer vision
OCR (Text Reader)
OCR reads letters from a picture and turns them into computer text.
Open try-it page →22 / Language and text models
Transformer
A Transformer reads many words at once and notices which words help explain other words.
Open try-it page →23 / Language and text models
BERT
BERT looks at the words before and after a word to understand what it means.
Open try-it page →24 / Language and text models
GPT
GPT writes text by guessing the next word again and again.
Open try-it page →25 / Language and text models
T5 (Text-to-Text)
T5 treats many language jobs as “text in, text out.”
Open try-it page →26 / Language and text models
Word2Vec
Word2Vec turns words into number points so words with similar meaning can be close together.
Open try-it page →27 / Audio and speech
Whisper
Whisper turns spoken audio into written words.
Open try-it page →28 / Audio and speech
Wav2Vec 2.0
Wav2Vec learns from sound by guessing missing parts of audio.
Open try-it page →29 / Audio and speech
CTC (Alignment)
CTC helps match speech sounds to text even when people speak at different speeds.
Open try-it page →30 / Neural networks and deep learning
RNN (Recurrent Net)
An RNN reads things in order and keeps a small memory of what just happened.
Open try-it page →31 / Neural networks and deep learning
LSTM (Long Memory)
An LSTM is an RNN with gates that help it remember important things for longer.
Open try-it page →32 / Image and content generation
Stable Diffusion
Stable Diffusion starts with messy dots and slowly cleans them into an image that matches a prompt.
Open try-it page →33 / Computer vision
CLIP
CLIP learns how pictures and words match each other.
Open try-it page →34 / Image and content generation
GANs
GANs use two AIs: one makes fake examples, and the other tries to spot the fake.
Open try-it page →35 / Image and content generation
CycleGAN
CycleGAN changes an image style, then checks if it can change it back.
Open try-it page →36 / Image and content generation
Deepfakes
Deepfake tools can copy face movement or appearance, so they must be used carefully.
Open try-it page →37 / Image and content generation
VAE (Autoencoders)
A VAE squeezes an image into a small code, adds a little randomness, and rebuilds a new image.
Open try-it page →38 / Computer vision
U-Net
U-Net finds exact shapes in an image by shrinking the image down and then building a detailed map back up.
Open try-it page →39 / Reinforcement learning, agents and planning
Q-Learning
Q-Learning learns by trying actions and remembering which ones lead to rewards.
Open try-it page →40 / Reinforcement learning, agents and planning
DQN
DQN is Q-Learning with a neural network that can look at a game screen.
Open try-it page →41 / Reinforcement learning, agents and planning
AlphaZero
AlphaZero learns board games by playing against itself many times.
Open try-it page →42 / Reinforcement learning, agents and planning
MCTS (Tree Search)
MCTS explores possible future moves by trying many pretend games.
Open try-it page →43 / Reinforcement learning, agents and planning
AutoGPT (Agents)
AutoGPT breaks a big goal into smaller tasks, tries them, checks the result, and improves its plan.
Open try-it page →44 / Reinforcement learning, agents and planning
A* Search
A* Search finds a path by looking at distance already traveled plus a guess of what is left.
Open try-it page →45 / Forecasting and optimization
ARIMA
ARIMA looks at past numbers in order and uses them to guess future numbers.
Open try-it page →46 / Forecasting and optimization
Prophet
Prophet forecasts numbers while paying attention to trends, seasons, and holidays.
Open try-it page →47 / Search and retrieval
TF-IDF
TF-IDF gives a word a high score when it is common in one document but rare in many others.
Open try-it page →48 / Search and retrieval
BM25
BM25 helps rank search results by checking how well a document matches the search words.
Open try-it page →49 / Search and retrieval
RAG (Open Book AI)
RAG lets an AI look up information first, then answer using what it found.
Open try-it page →50 / Language and text models
LDA
LDA groups documents by hidden topics, like sports, food, or science.
Open try-it page →51 / Forecasting and optimization
Survival Analysis
Survival analysis studies how long it may take before an event happens.
Open try-it page →52 / Neural networks and deep learning
GNN (Graph Networks)
A GNN learns from things and the connections between them.
Open try-it page →53 / Forecasting and optimization
Simplex Algorithm
Simplex finds the best choice when there are limits, like time, money, or materials.
Open try-it page →54 / Reinforcement learning, agents and planning
Multi-Armed Bandit
A multi-armed bandit balances trying new choices with using the choice that is winning so far.
Open try-it page →55 / Neural networks and deep learning
Neural Network
A neural network is a stack of simple math units that pass signals forward to make a guess.
Open try-it page →56 / Neural networks and deep learning
Deep Learning
Deep learning uses a neural network with many layers so it can learn harder patterns.
Open try-it page →57 / Image and content generation
Diffusion
Diffusion starts with noise and cleans it step by step into an image or video.
Open try-it page →58 / Reinforcement learning, agents and planning
AlphaGo
AlphaGo is a game-playing AI known for using search ideas to play the board game Go.
Open try-it page →No pages match. Try another word or group.
Every page uses plain language and real examples to show why each method matters.