Glossary of AI prompt terms and concepts
Are you new to the world of AI prompts and struggling to understand the terminology?
Have a peek at our comprehensive glossary, covering everything from the basics to the most complex concepts to help your understanding of the language and processes.
A
AI (Artificial Intelligence): The simulation of human intelligence in machines that are programmed to learn, reason, and self-correct.
B
Bot: A computer program designed to automate tasks, typically by simulating conversation with human users.
Byte: The basic unit of digital information, consisting of eight bits.
C
Chatbot: A type of bot designed to simulate conversation with human users through text or voice interactions.
Cloud Computing: The delivery of computing services, including servers, storage, databases, networking, software, analytics, and intelligence, over the Internet.
D
Deep Learning: A subset of machine learning that uses neural networks with many layers to learn from data.
Dialog System: A computer-based system that engages in dialog with humans, typically through natural language conversations.
E
Edge Computing: A distributed computing paradigm that brings computation and data storage closer to the location where it is needed, to improve response times and save bandwidth.
Encoder-Decoder: A type of neural network architecture that is commonly used for sequence-to-sequence tasks such as machine translation and summarization.
F
Fine-Tuning: The process of taking a pre-trained neural network and adapting it to a new task or domain by further training it on task-specific data.
Framework: A software library that provides a set of tools and abstractions for building applications.
G
GPU (Graphics Processing Unit): A specialized processor designed to handle the intense computational requirements of graphics rendering, but also used for accelerating machine learning and other compute-intensive workloads.
H
Hugging Face: A popular natural language processing library that provides pre-trained models and tools for building conversational AI applications.
Hyperparameter: A parameter that is set before training a machine learning model, such as the learning rate or the number of hidden layers.
I
Inference: The process of using a trained machine learning model to make predictions on new data.
Intent: In the context of natural language processing, an intent is the goal or purpose of a user’s input in a conversation.
J
Jailbreak: Jailbreaking is the process of modifying a prompt to remove the limitations imposed by the Ai. This allows the user to access additional features that are not permitted under standard prompting.
Jupyter Notebook: An open-source web application that allows users to create and share documents that contain live code, equations, visualizations, and narrative text.
K
Keras: A popular high-level neural network API written in Python, that is often used for rapid prototyping and experimentation.
Knowledge Graph: A type of database that captures the relationships between entities in a particular domain or knowledge area.
L
LSTM (Long Short-Term Memory): A type of recurrent neural network architecture that is commonly used for sequence modeling and prediction.
Lexicon: A list of words and their associated sentiment or emotion.
M
Machine Learning: A subset of artificial intelligence that focuses on building systems that can learn from and make decisions based on data, without being explicitly programmed.
Memory: A component of a neural network that stores information about previous inputs and outputs, and can influence the network’s current output.
N
Natural Language Processing (NLP): The field of study concerned with the interactions between computers and human languages, particularly how to program computers to process and analyze large amounts of natural language data.
Neural Network: A type of machine learning algorithm that is modeled after the structure and function of the human brain.
O
OpenAI: A research organization dedicated to developing and promoting friendly AI for the benefit of humanity.
OCR (Optical Character Recognition): The process of extracting text from images or scanned documents using machine learning algorithms.
P
Preprocessing: The process of cleaning and transforming raw data into a format that can be used for machine learning.
Pretrained models: AI models that have been trained on large amounts of data before being deployed. Pretrained models can be used as a starting point for further training or as a tool for various tasks, such as language generation or image recognition.
Q
Query: a question or request for information, often used in reference to a search engine or database.
R
Reinforcement Learning: a type of machine learning where an AI agent learns to make decisions based on rewards and punishments received from its environment.
S
Sentiment Analysis: a technique used in NLP to determine the emotional tone of a piece of text. It is often used to analyze customer feedback and reviews.
T
T is for Transfer Learning: a machine learning technique that involves applying knowledge learned from one task to another, often related, task.
U
Unsupervised Learning: a type of machine learning where an AI system learns to identify patterns in data without being given explicit labels or guidance.
V
Virtual Assistant: a software program designed to perform tasks or provide information for users, often using natural language processing.
W
Watson: an AI system developed by IBM that is known for its ability to process large amounts of unstructured data.
X
X is for eXplainable AI (XAI): a field of AI research focused on creating systems that can explain their decision-making processes to humans.
Y
Yield: a measure of the accuracy of an AI model, often expressed as the percentage of correct predictions.
Z
Zero-shot Learning: a machine learning technique where an AI system is able to learn to recognize new classes of objects or concepts without being explicitly trained on them. It achieves this by using knowledge learned from previously seen classes.
