Core Concepts
The field of computer science focused on creating systems capable of performing tasks that typically require human intelligence, including decision-making, visual recognition, language understanding, and problem-solving.
A branch of AI that uses advanced algorithms and machine learning models to create new content—such as text, images, music, or videos—by identifying and replicating patterns from existing data.
A subset of AI in which systems are trained to learn from data, identify patterns, and make decisions with minimal human intervention. ML includes supervised, unsupervised, and reinforcement learning techniques.
A field of AI that enables computers to understand, interpret, and respond to human language. Applications include chatbots, voice assistants, sentiment analysis, and automatic translation.
An advanced NLP model trained on massive text datasets to produce coherent and contextually appropriate text. Popular examples include OpenAI’s GPT series and Meta’s Llama 2.
Advanced AI Concepts
A computational model inspired by the human brain, consisting of interconnected nodes (“neurons”). Neural networks are foundational to many AI systems, including deep learning models.
A subset of machine learning that uses neural networks with multiple layers (“deep” architectures) to perform complex tasks such as image recognition, natural language understanding, and autonomous driving.
An ML approach where an agent learns to make decisions by interacting with an environment, receiving rewards for desirable outcomes, and penalties for undesirable ones.
Techniques and methods that make AI models more transparent and their decision-making processes easier to understand for humans, addressing ethical and accountability concerns.
A growing field examining the societal impact of AI, focusing on issues such as bias, fairness, transparency, accountability, and the ethical use of AI tools.
Practical AI Terminology
An AI-driven tool that uses NLP to simulate human-like conversations via text or voice interfaces, often used in customer service, education, or entertainment.
A term describing when an AI model generates output that is fabricated or incorrect, often due to limitations in training data or overfitting.
A directive or query provided to an AI system to generate a specific response. Effective prompts are often clear, concise, and contextually detailed.
The art and science of refining prompts to optimize AI performance and generate more accurate or useful results.
The dataset used to teach an AI system by exposing it to patterns, relationships, and examples. High-quality, diverse training data is critical for robust AI models.
A situation where an AI model performs well on its training data but fails to generalize to new or unseen data, often due to excessive complexity or inadequate data diversity.
The opposite of overfitting, where a model is too simplistic to capture the underlying patterns in the data, which leads to poor performance on both training and new data.
Model and System Parameters
Adjustable factors within an AI model, such as weights and biases, which are tuned during training to optimize performance.
A parameter in generative AI models that adjusts the randomness of generated outputs. Higher values increase variability, while lower values produce more deterministic responses.
The smallest units of text (e.g., words or subwords) that AI models process. For example, the phrase “artificial intelligence” might be split into three tokens: “artificial,” “intelligence,” and a space.
The process of retraining a pre-trained model on specific data to adapt it to a particular task or domain.
The phase when an AI model applies what it has learned during training to make predictions or generate responses based on new input data.
Emerging Technologies and Tools
An AI field focused on enabling machines to interpret and understand visual information from the world, such as recognizing objects, faces, or actions in images and videos.
A privacy-preserving ML approach where models are trained across multiple decentralized devices using local data, without transferring that data to a central server.
A capability of AI systems to perform tasks or recognize concepts without having been explicitly trained on them, leveraging general knowledge learned from other tasks.
Guidelines and frameworks adopted by organizations to govern the responsible use of AI, ensuring alignment with societal values and regulatory requirements.
Note: These definitions were generated using chatGPT 4.o and have been reviewed for accuracy by Texas State University Libraries staff. When using content generated by large language models, careful review and proper attribution are essential.