Chapter 1, Introduction to One-shot Learning, tells us what one-shot learning is and how it works. It also tells us about the human brain's workings and how it translates to machine learning.
Chapter 2, Metrics-Based Methods, explores methods that use different forms of embeddings, and evaluation metrics, by keeping the core as basic k-nearest neighbors.
Chapter 3, Model-Based Methods, explores two architectures whose internal architectures help to train a k-shot learning model.
Chapter 4, Optimization-Based Methods, explores different forms of optimization algorithms, which help in improving accuracy even when the volume of data is low.
Chapter 5, Generative Modeling-Based Methods, explores the development of a Bayesian learning framework based on representing object categories with probabilistic models.
Chapter 6, Conclusions and Other Approaches, goes through certain aspects of architecture, metrics, and algorithms to understand how we can determine whether an approach is close to human brain capability.
