AI Starter Guide

This book explains one topic per page, like a big glossary, easy wiki, quick encyclopedia, or summary notes. Edited by Joel Parker Henderson (@joelparkerhenderson).

Contents

Introduction

• What is this book?

• Who is this for?

• Why am I creating this?

• Are there more guides?

Introduction

Artificial intelligence

• Artificial General Intelligence (AGI)

• Artificial Super Intelligence (ASI)

• Natural language processing (NLP)

• Explainable Artificial Intelligence (XAI)

• Symbolic artificial intelligence

• Expert system

TODO

• AI agent

• AI alignment

• AI ethics

• Chain of thought

• AI hallucination

• Chatbot

• Neural Radiance Fields (NeRF)

• Hyperparameter tuning

• Machine learning parameters

• dendrogram

AI datasets

• Training data

• Validation data

• Test data

Computer processors

• Central Processing Unit (CPU)

• Graphics Processing Unit (GPU)

• Tensor Processing Unit (TPU)

• Vision Processing Unit (VPU)

• AI processor

• Field Programmable Gate Array (FPGA)

Machine learning

• Supervised learning

• Unsupervised learning

• Reinforcement learning

• Deep learning

• Backpropagation

• Forward Propagation

• Gradient descent

• Zero-shot learning

• Hidden Markov Model (HMM)

Machine learning algorithms

• Decision Tree

• instance-based learning

• lazy learning algorithms

Supervised

Supervised learning algorithms

• Support Vector Machine (SVM)

• Linear Regression: Used for predicting continuous numerical values.

• Logistic Regression: Used for binary classification problems.

• Decision Trees: Tree-based models for both classification and regression tasks.

• Random Forest: An ensemble method combining multiple decision trees.

Unsupervised

Unsupervised learning algorithms

• Self-Organizing Maps (SOM)

• Kohonen maps → Self-Organizing Maps (SOM)

Clustering algorithms

• K-means Clustering: Partition data points into k clusters based on their proximity to cluster centroids.

• Hierarchical clustering

Dimensionality reduction algorithms

• Principal Component Analysis (PCA)

• t-Distributed Stochastic Neighbor Embedding (t-SNE)

Anomaly detection algorithms

Statistical Methods:

• Modified Z-Score

• Z-Score

• Percentile: This method identifies anomalies based on percentiles or quantiles of the data distribution.

Density-Based Methods:

• Local Outlier Factor (LOF)

• Isolation Forest

• Density-Based Spatial Clustering of Applications with Noise (DBSCAN)

Proximity-Based Methods:

• k-Nearest Neighbors (KNN)

• Distance-Based Outlier Detection (LOCI)

Machine Learning-Based Methods:

• Autoencoders

• One-Class Support Vector Machines (One-Class SVM)

Ensemble Methods:

• Majority voting

• Isolation Forest Ensemble

Generative models

• Gaussian Mixture Models (GMM): Model a combination of distributions, allowing data generation and density estimation.

• Autoencoders: Neural networks used for unsupervised feature learning.

• Variational Autoencoders (VAE): Learn to generate new data samples by mapping them to a latent space.

Ensemble learning algorithms

• Bagging (a.k.a. Bootstrap Aggregating)

• Random Forest

• Boosting

• Gradient Boosting Machines (GBM)

• Extreme Gradient Boosting (XGBoost)

• LightGBM

• Stacking (a.k.a. Stacked Generalization)

• Voting Classifiers (a.k.a. Voting Ensembles)

Unsupervised learning tasks

• Clustering

• Dimensionality Reduction

• Anomaly Detection

• Density Estimation

Large Language Model (LLM)

• Generative Pretrained Transformer (GPT)

• Contrastive Language-Image Pretraining (CLIP)

Neural Network (NN)

• Convolutional Neural Network (CNN)

• General Adversarial Network (GAN)

• Recurrent Neural Network (RNN)

• Deep Neural Network (DNN)

• Transformer architecture

Activation function

• Hyperbolic Tangent (tanh) activation function

• Rectified Linear Unit (ReLU) activation function

• Sigmoid activation function

Loss function

• Mean Squared Error (MSE)

• Mean Absolute Error (MAE)

Kernel trick

• linear kernel

• polynomial kernel

• radial basis function (RBF) kernel

• sigmoid kernel

Machine learning performance metrics

• Accuracy

• Precision

• Recall (≡ True Positive Rate, Sensitivity)

• Specificity (≡ True Negative Rate)

• F1-Score

• Receiver Operating Characteristic (ROC)

• Area Under the Curve (AUC)

• R-squared (R2)

• Silhouette Score

• Davies-Bouldin Index

• Adjusted Rand Index (ARI)

• Overfitting

• Underfitting

• True Positive Rate → Recall

• True Negative Rate → Specificity

AI tools

• AI content generator

• AI image generation

• AI form fill

• AI UI/UX

• AI internationalization/localization

• AI plagiarism checker

AI for business areas

• AI sales

• AI marketing

• AI accounting

• AI human resources

• AI resource leveling

• AI customer service

• AI for business strategy

• AI for partner management

• AI for product development

• AI for project management

• AI for software programming

AI + business sectors

• AI + adtech (advertising tech)

• AI + agtech (agricultural tech)

• AI + biotech (biological tech)

• AI + cleantech (clean energy tech)

• AI + edtech (educational tech)

• AI + fintech (financial tech)

• AI + govtech (governmental tech)

• AI + legtech (legal tech)

• AI + martech (marketing tech)

• AI + medtech (medical tech)

• AI + realtech (real estate tech)

• AI + regtech (regulatory tech)

Conclusion

• About the editor

• About the AI

• About the ebook