Ranking Distance Calibration For Cross-domain Few-shot Learning
2021 · Pan Li, Shaogang Gong, Chengjie Wang, et al.
Abstract
Recent progress in few-shot learning promotes a more realistic cross-domain setting, where the source and target datasets are from different domains. Due to the domain gap and disjoint label spaces between source and target datasets, their shared knowledge is extremely limited. This encourages us to explore more information in the target domain rather than to overly elaborate training strategies on the source domain as in many existing methods. Hence, we start from a generic representation pre-trained by a cross-entropy loss and a conventional distance-based classifier, along with an image retrieval view, to employ a re-ranking process for calibrating a target distance matrix by discovering the reciprocal k-nearest neighbours within the task. Assuming the pre-trained representation is biased towards the source, we construct a non-linear subspace to minimise task-irrelevant features therewithin while keep more transferrable discriminative information by a hyperbolic tangent transformati
Authors
(none)
Tags
Stats
Related papers
- Finding Significant Features For Few-shot Learning Using Dimensionality Reduction (2021)2.26
- P3dc-shot: Prior-driven Discrete Data Calibration For Nearest-neighbor Few-shot Classification (2023)6.34
- Susana Distancia Is All You Need: Enforcing Class Separability In Metric Learning Via Two Novel Distance-based Loss Functions For Few-shot Image Classification (2023)0.00
- Improving Calibration In Deep Metric Learning With Cross-example Softmax (2020)0.00
- Gmm-based Comprehensive Feature Extraction And Relative Distance Preservation For Few-shot Cross-modal Retrieval (2025)2.26
- Rankdnn: Learning To Rank For Few-shot Learning (2022)12.58
- Few-shot Metric Learning: Online Adaptation Of Embedding For Retrieval (2022)8.09
- Feature Representation Learning For Unsupervised Cross-domain Image Retrieval (2022)11.46