1 code implementation • 6 Jun 2021 • John Cai, Bill Cai, ShengMei Shen
In this paper, we propose Domain Agnostic Meta Score-based Learning (DAMSL), a novel, versatile and highly effective solution that delivers significant out-performance over state-of-the-art methods for cross-domain few-shot learning.
Ranked #1 on Cross-Domain Few-Shot on miniImagenet
no code implementations • 3 Dec 2020 • John Cai, Bill Cai, Sheng Mei Shen
Our method, called Score-based Meta Transfer-Learning (SB-MTL), combines transfer-learning and meta-learning by using a MAML-optimized feature encoder and a score-based Graph Neural Network.
no code implementations • 21 May 2020 • John Cai, Sheng Mei Shen
In our final results, we combine the novel method with the baseline method in a simple ensemble, and achieve an average accuracy of 73. 78% on the benchmark.