Don't miss an insight. Subscribe to Techopedia for free.


Zero-Shot, One-Shot, Few-Shot Learning

What Does Zero-Shot, One-Shot, Few-Shot Learning Mean?

Zero-shot learning, few-shot learning and one-shot learning are all techniques that allow a machine learning model to make predictions for new classes with limited labeled data. The choice of technique depends on the specific problem and the amount of labeled data available for new categories or labels (classes).


One-shot learning – each new class has one labeled example. The goal is to make predictions for the new classes based on this single example.

Few-shot learning – there is a limited number of labeled examples for each new class. The goal is to make predictions for new classes based on just a few examples of labeled data.

Zero-shot learning – there is absolutely no labeled data available for new classes. The goal is for the algorithm to make predictions about new classes by using prior knowledge about the relationships that exist between classes it already knows. In the case of large language models (LLMs) like ChatGPT, for example, prior knowledge is likely include semantic similarities.

Techopedia Explains Zero-Shot, One-Shot, Few-Shot Learning

Zero-shot, few-shot and one-shot learning are important concepts in AI research because when executed successfully, they allow AI systems to be more flexible, scalable and effective in real-world scenarios.

Different approaches to zero-shot, few-shot and one-shot learning include:

Attribute-based approaches – the model uses relationships between attributes to generalize its knowledge and apply the knowledge to new classes instead of relying on labeled examples.

Embedding-based approaches — the model infers information about new classes based on their proximity to known classes in the embedding space.

Generative approaches — the model generates synthetic examples for unseen categories based on their semantic representation.

Metric-based models – the model learns a similarity metric between features of the input data and the features of each class and then uses this metric to make predictions for new, unseen classes.

Neural network-based models – the model uses convolutional neural networks (CNNs) or recurrent neural networks (RNNs) to learn about the correlations between input data and class predictions.

Transfer learning-based models – the model is pre-trained with immense amounts of general-purpose training data and then fine-tuned with targeted labeled data for a specific task.

Importance of Zero-Shot, One-Shot and Few-Shot Learning

In many real-world scenarios, it is not feasible to collect and label large amounts of data for every possible class or concept that a model may encounter. Allowing models to handle new and unseen classes with limited or no additional labeled data can improve scalability and help reduce the costs associated with labeling and annotating data.


Related Terms