GH-ESD: Grounded Hypothesis-Driven Error Slice Discovery for Instance-Level Vision Tasks
Authors Wei Zhang*, Chaoqun Wang*, Zixuan Guan, Ping Sheng Kao, Pengfei Zhao, Peng Wu, Sifeng He
View publication：https://arxiv.org/abs/2512.24592
Instance-Level Task Parameters: A Robust Multi-task Weighting Framework
June 24, 2021 research area Computer Vision：/research/?domain=Computer%20Vision Workshop at CVPR：/research/?event=CVPR%20Workshop
Recent works have shown that deep neural networks benefit from multi-task learning by learning a shared representation across several related tasks. However, performance of such systems depend on relative weighting between various losses involved during training. Prior works on loss weighting schemes assume that instances are equally easy or hard for all tasks. In order to break this assumption, we let the training process dictate the optimal…
Sliced Wasserstein Discrepancy for Unsupervised Domain Adaptation
March 10, 2019 research area Methods and Algorithms：/research/?domain=Methods%20and%20Algorithms conference CVPR：/research/?event=CVPR
In this work, we connect two distinct concepts for unsupervised domain adaptation: feature distribution alignment between domains by utilizing the task-specific decision boundary and the Wasserstein metric. Our proposed sliced Wasserstein discrepancy (SWD) is designed to capture the natural notion of dissimilarity between the outputs of task-specific classifiers. It provides a geometrically meaningful guidance to detect target samples that are…
Our research in machine learning breaks new ground every day.
Scan to open this article