Abstract:Despite the widespread success of deep learning, its intense requirements for vast amounts of data and extensive training make it impractical for various real-world applications where data is scarce. In recent years, Few-Shot Learning (FSL) has emerged as a learning paradigm that aims to address these limitations by leveraging prior knowledge to enable rapid adaptation to novel learning tasks. Due to its properties that highly complement deep learning's data-intensive needs, FSL has seen significant growth in the past few years. This survey provides a comprehensive overview of both well-established methods as well as recent advancements in the FSL field. The presented taxonomy extends previously proposed ones by incorporating emerging FSL paradigms, such as in-context learning, along with novel categories within the meta-learning paradigm for FSL, including neural processes and probabilistic meta-learning. Furthermore, a holistic overview of FSL is provided by discussing hybrid FSL approaches that extend FSL beyond the typically examined supervised learning setting. The survey also explores FSL's diverse applications across various domains. Finally, recent trends shaping the field, outstanding challenges, and promising future research directions are discussed.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is: **How to achieve effective machine - learning model training in the case of scarce data, especially for the Few - Shot Learning (FSL) problem.** Specifically, although modern deep - learning models have achieved remarkable success in many fields, their dependence on large amounts of data and computing resources makes them face challenges in practical applications, especially in cases where data is scarce or difficult to obtain. Few - Shot Learning aims to reduce the need for large amounts of data and improve the generalization ability of models in the case of a small number of samples by using prior knowledge to quickly adapt to new tasks. The main contributions of the paper include: 1. **Proposing a new taxonomy**: Expanding the existing classification framework and introducing new sub - categories, such as neural processes and probabilistic meta - learning. 2. **Analyzing In - Context Learning for the first time**: Exploring its similarities and differences with traditional FSL methods and various techniques for combining these two paradigms. 3. **Exploring the connection between FSL and other related learning problems**: Reviewing existing hybrid methods that extend the standard supervised FSL framework to different settings, such as unsupervised FSL, federated FSL, cross - domain FSL, etc. 4. **Highlighting the latest trends and challenges in the FSL field**: Pointing out the key issues that need to be resolved to promote the development of this field, such as human - like learning and green AI. Through these efforts, the paper provides a comprehensive and up - to - date review of Few - Shot Learning, covering a wide range of content from classical methods to emerging paradigms, and discusses future research directions.

A Complete Survey on Contemporary Methods, Emerging Paradigms and Hybrid Approaches for Few-Shot Learning

A Comprehensive Survey of Few-shot Learning: Evolution, Applications, Challenges, and Opportunities

Generalizing from a Few Examples: A Survey on Few-Shot Learning

Deep Learning for Cross-Domain Few-Shot Visual Recognition: A Survey

Meta-learning Approaches for Few-Shot Learning: A Survey of Recent Advances

A survey on machine learning from few samples

Learning from Very Few Samples: A Survey.

A Survey of Few-Shot Learning: An Effective Method for Intrusion Detection

Few-Shot Object Detection: A Comprehensive Survey

Learning from Few Examples: A Summary of Approaches to Few-Shot Learning

A Survey of Few-Shot Learning on Graphs: from Meta-Learning to Pre-Training and Prompt Learning

Towards Few-Shot Learning in the Open World: A Review and Beyond

An Overview of Deep Learning Architectures in Few-Shot Learning Domain

A survey on few-shot class-incremental learning

Zero-shot and Few-shot Learning with Knowledge Graphs: A Comprehensive Survey

Few-Shot Image Classification: Current Status and Research Trends

Semantic-based Selection, Synthesis, and Supervision for Few-shot Learning

A Survey of Self-Supervised and Few-Shot Object Detection

Skeletal muscle changes in scurvy with a note on the mechanism of the attachment of myofibrils to tendon.