Loading...
51 terms across 9 categories. Explore key concepts in artificial intelligence and machine learning.
Ensuring AI systems behave in accordance with human values and intentions.
Research and practices aimed at ensuring AI systems are safe and beneficial.
The simulation of human intelligence in machines programmed to think and learn.
A technique that allows models to focus on relevant parts of the input when producing output.
The algorithm used to train neural networks by computing gradients of the loss.
Systematic errors in AI systems that create unfair outcomes for certain groups.
A prompting technique that encourages models to show reasoning steps before conclusions.
AI field enabling machines to interpret and understand visual information from images and videos.
The maximum amount of text (tokens) a model can process in a single interaction.
Neural networks designed for processing grid-like data, especially images.
Changes in data distribution over time that can degrade model performance.
A subset of ML using neural networks with many layers to learn complex patterns.
Generative models that create data by learning to reverse a gradual noising process.
Dense vector representations of data that capture semantic meaning.
The ability to understand and explain how AI models make their decisions.
Teaching a model new tasks by providing just a few examples in the prompt.
Adapting a pre-trained model to a specific task or domain with additional training.
Two neural networks competing to generate realistic synthetic data.
AI systems that can create new content including text, images, audio, code, and video.
An optimization algorithm that iteratively adjusts parameters to minimize the loss function.
When AI models generate plausible-sounding but incorrect or fabricated information.
The task of assigning a label or category to an entire image.
Partitioning an image into multiple segments or regions at the pixel level.
AI models trained on massive text datasets to understand and generate human-like text.
A subset of AI that enables systems to learn and improve from experience without being explicitly programmed.
Practices and tools for deploying and maintaining ML models in production reliably.
The process of making a trained ML model available for use in production.
Tracking model performance and behavior in production to detect issues.
AI systems that can process and generate multiple types of data like text, images, and audio.
Identifying and classifying named entities (people, places, organizations) in text.
AI field focused on enabling computers to understand, interpret, and generate human language.
A computing system inspired by biological neural networks that can learn to recognize patterns.
Technique that applies the artistic style of one image to the content of another.
Identifying and locating multiple objects within an image with bounding boxes.
Technology that converts images of text into machine-readable text data.
When a model performs well on training data but poorly on new, unseen data.
The practice of crafting effective inputs to get desired outputs from AI models.
ML paradigm where agents learn to make decisions by receiving rewards or penalties.
A technique that enhances LLM responses by retrieving relevant information from external sources.
Training AI using human preferences to guide reinforcement learning.
Search that understands meaning and intent, not just keyword matching.
The process of determining the emotional tone or opinion expressed in text.
A type of ML where the model is trained on labeled data with known correct answers.
Automatically categorizing text into predefined groups or labels.
AI systems that create images from natural language descriptions.
The process of breaking text into smaller units (tokens) for model processing.
Using knowledge from one task to improve learning on a different but related task.
A neural network architecture that uses self-attention mechanisms, revolutionizing NLP.
ML approach where the model finds patterns in data without labeled examples.
Databases optimized for storing and querying high-dimensional embedding vectors.
Transformer architecture adapted for image understanding by treating images as sequences of patches.