Acronym soup forArtificial Intelligence
AI enables machines to mimic human intelligence, solving tasks like learning, decision-making, and language understanding. It powers innovations across industries, using tools like machine learning, NLP, and computer vision to improve efficiency and unlock possibilities.
An open protocol referring to direct communication, negotiation, or collaboration between autonomous AI agents.
A type of AI capable of understanding, learning, and performing tasks at a human level.
The simulation of human intelligence in machines that perform tasks like learning and reasoning.
A type of ANN designed for processing structured grid data like images.
A type of ANN where connections do not form cycles.
A type of LLM designed to generate coherent and contextually relevant text.
A machine learning model designed for natural language processing tasks such as language generation. These language models use many parameters trained with self-supervised learning on vast textual data. In the early 2020s, several LLMs became multimodal, processing and generating different data types, such as images or audio.
A dimensionality reduction technique often used in AI preprocessing.
A technique used to align AI behavior with human preferences by incorporating feedback.
A process combining supervised pre-training with reinforcement learning to refine AI behavior using feedback. It's used to align models like chatbots with specific goals or ethical guidelines by adjusting outputs via methods like PPO.
A supervised learning model for classification and regression.