AI Glossary

artificial intelligence (AI): Any variety of system that performs tasks that are traditionally thought by humans to require intelligence.

computer vision (CV): The field of AI methods that process visual data such as images or videos.

data augmentation: The process of artificially increasing the size of a dataset by adding new examples that are somehow generated from existing ones.

hyper-parameter: A parameter that is not learned by the model, but is instead specified by the user. For example, the number of layers in a neural network, or the learning rate used when optimizing the model.

machine learning (ML): The process of learning to make predictions or perform actions based on data.

machine translation (MT): A natural language processing task that involves translating between one human language and another.

multi-task learning: A method of training a model to perform multiple tasks at once, rather than training a separate model for each task.

natural language processing (NLP): The field of AI methods that process human languages such as English or Chinese.

semi-supervised learning: Learning based on a combination of supervised learning and unsupervised learning.

supervised learning: Learning in the case where both inputs and gold-standard outputs are provided during the learning process.

unsupervised learning: Learning in the case where no pairs of inputs and gold-standard outputs are provided.

prompt engineering: A machine learning paradigm that focuses on designing a good prompt to better elicit knowledge from large pre-trained models.