AI Terms

Artificial Intelligence

AI is the simulation of human intelligence in machines that are programmed to think and learn like humans. AI systems can perform tasks that typically require human intelligence, such as visual perception, speech recognition, decision-making, and language translation.

AI Model

An AI model is a program trained to recognize patterns in data and make decisions or predictions.

AI Prompt

An AI prompt is input text or data given to an AI system to instruct it to perform a specific task or generate a specific kind of output.

Algorithm

A set of rules or instructions given to a computer to help it learn on its own and make decisions or solve problems.

Artificial General Intelligence (AGI)

A hypothetical form of AI that would have the ability to understand, learn, and apply knowledge across a wide range of tasks at a level equal to or beyond human capability.

Backpropagation

A method used in training neural networks where errors are sent backwards through the network to adjust weights and improve accuracy.

Bias (AI)

Systematic errors or unfairness in AI systems that can lead to discriminatory outcomes, often reflecting biases present in training data.

Big Data

Extremely large datasets that may be analyzed computationally to reveal patterns, trends, and associations, especially relating to human behavior and interactions. Big Data is crucial for training effective AI models.

ChatGPT

ChatGPT is an AI chatbot developed by OpenAI that uses large language models to generate responses. Gemini is Google's family of AI models for tasks similar to ChatGPT.

Classification

A type of machine learning task that assigns input data to predefined categories or classes, such as spam detection or image recognition.

Computer Vision

A field of AI that enables computers to interpret and understand visual information from images and videos, similar to human vision.

Data Mining

The process of discovering patterns, correlations, and insights from large datasets using statistical and computational methods.

Deep Learning

Deep learning is a type of machine learning that uses neural networks with many layers to model complex patterns in large amounts of data. DeepSeek is a family of open-source large language models that apply deep learning techniques to text generation and understanding.

Fine-tuning

The process of taking a pre-trained model and adapting it to a specific task by training it further on task-specific data.

Generative AI

Generative AI refers to systems that can create new content such as text, images, audio, or video using machine learning models trained on large datasets. The name "generative" means they can generate new outputs that aren't just copies of the input. They excel at handling sequential data, especially in Natural Language tasks.

Generative Adversarial Network (GAN)

A class of machine learning frameworks where two neural networks contest with each other to produce data that is indistinguishable from real data.

Hallucination

In the AI world, a hallucination is basically a lie, something the AI got wrong. The AI is designed to create output, and so it will, good or bad. You should verify everything you can about the AI's answer. You can ask it to provide its sources, or explain its reasoning.

LLM (Large Language Model)

A type of AI model trained on vast amounts of text data to understand and generate human-like language. These models can perform various language tasks such as translation, summarization, and conversation.

Machine Learning (ML)

A subset of AI that enables computers to learn and improve from experience without being explicitly programmed. Machine learning algorithms build mathematical models based on training data to make predictions or decisions.

Natural Language Processing (NLP)

A branch of AI that helps computers understand, interpret, and generate human language in a valuable way.

Neural Network

A neural network is a computing system inspired by the human brain's structure, used for recognizing patterns and making decisions. These networks can learn to recognize patterns and make decisions by adjusting the strength of connections between neurons. A neural net can "learn" from its experience. For example, you train a neural net that plays Tic-Tac-Toe by playing games against it, and it learns how to play.

Overfitting

A modeling error that occurs when a function is too closely aligned to a limited set of data points. This can cause the model to perform well on training data but poorly on new, unseen data.

Prompt Engineering

The practice of designing and refining input prompts to effectively communicate with AI systems, particularly language models, to achieve desired outputs or behaviors.

Prompt Template

A prompt template is a structured format used to generate consistent and useful prompts for AI systems.

RAG (Retrieval Augmented Generation)

A technique that enhances the accuracy and reliability of generative AI models by incorporating information from specific and relevant data sources. It allows large language models (LLMs) to access external knowledge bases, ensuring that their responses are grounded in up-to-date and accurate information. This approach helps mitigate issues like hallucinations, where models generate plausible but incorrect answers, by providing sources that can be cited for verification.

Regression

A type of machine learning task that predicts continuous numerical values, such as predicting house prices or stock values.

Reinforcement Learning

A type of machine learning where an agent learns to make decisions by performing actions in an environment and receiving rewards or penalties. The system learns through trial and error to maximize cumulative rewards.

Supervised Learning

A type of machine learning where algorithms learn from labeled training data, meaning the input data comes with known correct answers. The system learns to map inputs to outputs based on these examples.

Test Data

A dataset held out from training used to evaluate the final performance of a machine learning model.

Tokenization

The process of breaking down text into smaller units (tokens) such as words or subwords for processing by AI models.

Training Data

The dataset used to teach a machine learning algorithm, containing input examples and their corresponding correct outputs.

Transfer Learning

A technique where a model trained on one task is adapted for a related task, leveraging previously learned knowledge.

Transformer

A Transformer is a type of deep learning model that excels at handling sequential data, especially in Natural Language tasks. Note that the last "T" in "ChatGPT" stands for Transformer.

Underfitting

A modeling error that occurs when a model is too simple to capture the underlying patterns in the data. This leads to poor performance on both training and new data.

Unsupervised Learning

A machine learning approach where algorithms find patterns in data without being given labeled examples. The system identifies hidden structures in data where the correct answers are not provided.

Validation Data

A separate dataset used to evaluate model performance during training and tune hyperparameters without overfitting.

Weights

Numerical values in a neural network that determine the strength of connections between neurons and are adjusted during training.

AI Course Terms

AI Terms to Know