LGJul 29, 2022

A Survey of Learning on Small Data: Generalization, Optimization, and Challenge

arXiv:2207.14443v23 citationsh-index: 72
Originality Synthesis-oriented
AI Analysis

It addresses the problem of high annotation and training costs in AI by surveying methods for learning on small data, but it is incremental as it reviews existing topics without presenting new empirical results.

This survey analyzes the generalization error and label complexity of learning on small data using a PAC framework, and introduces geometric representation perspectives like Euclidean and non-Euclidean means to discuss optimization solutions.

Learning on big data brings success for artificial intelligence (AI), but the annotation and training costs are expensive. In future, learning on small data that approximates the generalization ability of big data is one of the ultimate purposes of AI, which requires machines to recognize objectives and scenarios relying on small data as humans. A series of learning topics is going on this way such as active learning and few-shot learning. However, there are few theoretical guarantees for their generalization performance. Moreover, most of their settings are passive, that is, the label distribution is explicitly controlled by finite training resources from known distributions. This survey follows the agnostic active sampling theory under a PAC (Probably Approximately Correct) framework to analyze the generalization error and label complexity of learning on small data in model-agnostic supervised and unsupervised fashion. Considering multiple learning communities could produce small data representation and related topics have been well surveyed, we thus subjoin novel geometric representation perspectives for small data: the Euclidean and non-Euclidean (hyperbolic) mean, where the optimization solutions including the Euclidean gradients, non-Euclidean gradients, and Stein gradient are presented and discussed. Later, multiple learning communities that may be improved by learning on small data are summarized, which yield data-efficient representations, such as transfer learning, contrastive learning, graph representation learning. Meanwhile, we find that the meta-learning may provide effective parameter update policies for learning on small data. Then, we explore multiple challenging scenarios for small data, such as the weak supervision and multi-label. Finally, multiple data applications that may benefit from efficient small data representation are surveyed.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes