8 terms
Showing all terms starting with O
When a model learns the training data too closely, including its noise, and fails to generalise to unseen data.
AI models or tools whose source code and/or weights are publicly available, allowing anyone to use, study, and modify them.
The coordination of multiple AI models, tools, and data sources in a pipeline to complete complex multi-step tasks.
A computer vision task that identifies and localises objects within an image by drawing bounding boxes around each detected instance.
Running AI models directly on end-user hardware such as smartphones or laptops, enabling offline use and enhanced data privacy.
Training or prompting an AI to learn from a single example, useful when labelled data is extremely scarce.
A formal representation of knowledge within a domain, defining concepts, relationships, and rules - often used to enhance AI reasoning.
The final layer of a neural network that produces the model's predictions, formatted according to the task (classification, generation, etc.).
