The Importance of Data Lineage for Reliable Data Products

Short Learning and Zero Shot Learning in AI

Short learning and zero shot learning are two different approaches in artificial intelligence that are used to train machine learning models.

Short Learning

Short learning is a type of machine learning where the model is trained on a small amount of data. This approach is useful when there is limited data available or when the data is expensive to collect. Short learning algorithms are designed to quickly learn from a small amount of data and make accurate predictions.

Short learning algorithms include techniques such as transfer learning, where a pre-trained model is used as a starting point for a new task, and few-shot learning, where the model is trained on a small number of examples for each class.

Zero Shot Learning

Zero shot learning is a type of machine learning where the model is trained to recognize objects or concepts that it has never seen before. This approach is useful when there are many different classes to recognize and it is impractical to collect data for all of them.

Zero shot learning algorithms use semantic embeddings to represent objects or concepts in a high-dimensional space. The model is trained to recognize these embeddings and can then recognize new objects or concepts based on their similarity to the embeddings.

When to Use Short Learning and Zero Shot Learning

Short learning is useful when there is limited data available or when the data is expensive to collect. It is also useful when the model needs to be trained quickly or when the model needs to be updated frequently.

Zero shot learning is useful when there are many different classes to recognize and it is impractical to collect data for all of them. It is also useful when the model needs to recognize new objects or concepts that it has never seen before.