VALSE 论文速览 第167期：BuresNet：基于条件Bures度量的可迁移表征学习 ...

论文题目：

BuresNet: Conditional Bures Metric for Transferable Representation Learning

作者列表：

任传贤 (中山大学), 罗又维 (中山大学), 戴道清 (中山大学)

B站观看网址：

https://www.bilibili.com/video/BV1SK421e7vP/

论文摘要：

As a fundamental manner for learning and cognition, transfer learning has attracted widespread attention in recent years. Typical transfer learning tasks include unsupervised domain adaptation (UDA) and few-shot learning (FSL), which both attempt to sufficiently transfer discriminative knowledge from the training environment to the test environment to improve the model's generalization performance. Previous transfer learning methods usually ignore the potential conditional distribution shift between environments. This leads to the discriminability degradation in the test environments. Therefore, how to construct a learnable and interpretable metric to measure and then reduce the gap between conditional distributions is very important in the literature. In this article, we design the Conditional Kernel Bures (CKB) metric for characterizing conditional distribution discrepancy, and derive an empirical estimation with convergence guarantee. CKB provides a statistical and interpretable approach, under the optimal transportation framework, to understand the knowledge transfer mechanism. It is essentially an extension of optimal transportation from the marginal distributions to the conditional distributions. CKB can be used as a plug-and-play module and placed onto the loss layer in deep networks, thus, it plays the bottleneck role in representation learning. From this perspective, the new method with network architecture is abbreviated as BuresNet, and it can be used extract conditional invariant features for both UDA and FSL tasks. BuresNet can be trained in an end-to-end manner. Extensive experiment results on several benchmark datasets validate the effectiveness of BuresNet.

参考文献：

[1] IEEE Transactions on Pattern Analysis and Machine Intelligence (Volume: 45, Issue: 4, 01 April 2023).

论文链接：

[https://ieeexplore.ieee.org/document/9829324]

代码链接：

[https://github.com/MathAI-LAB/CKB]

视频讲者简介：

Chuan-Xian Ren is currently a professor with the School of Mathematics, Sun YatSen University. During 2010 and 2011, he was with the Department of Electronic Engineering, City University of Hong Kong, as a senior research associate. His research interests include visual intelligence based pattern analysis and machine learning.

特别鸣谢本次论文速览主要组织者：

月度轮值AC：张瑞茂 (香港中文大学 (深圳))