CVMar 31, 2020

DPGN: Distribution Propagation Graph Network for Few-shot Learning

arXiv:2003.14247v2235 citations
AI Analysis

This addresses the problem of learning from limited data for AI systems, offering a novel hybrid method that is incremental but shows strong gains.

The paper tackles few-shot learning by proposing DPGN, a graph network that models both distribution-level and instance-level relations, achieving state-of-the-art results with improvements of 5-12% in supervised and 7-13% in semi-supervised settings.

Most graph-network-based meta-learning approaches model instance-level relation of examples. We extend this idea further to explicitly model the distribution-level relation of one example to all other examples in a 1-vs-N manner. We propose a novel approach named distribution propagation graph network (DPGN) for few-shot learning. It conveys both the distribution-level relations and instance-level relations in each few-shot learning task. To combine the distribution-level relations and instance-level relations for all examples, we construct a dual complete graph network which consists of a point graph and a distribution graph with each node standing for an example. Equipped with dual graph architecture, DPGN propagates label information from labeled examples to unlabeled examples within several update generations. In extensive experiments on few-shot learning benchmarks, DPGN outperforms state-of-the-art results by a large margin in 5% $\sim$ 12% under supervised setting and 7% $\sim$ 13% under semi-supervised setting. Code will be released.

Code Implementations1 repo
Foundations

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

Your Notes