Picture for Hong Xuan

Hong Xuan

Dissecting the impact of different loss functions with gradient surgery

Jan 27, 2022
Hong Xuan, Robert Pless

Access Paper or Ask Questions

Hard negative examples are hard, but useful

Jul 24, 2020
Hong Xuan, Abby Stylianou, Xiaotong Liu, Robert Pless

Figure 1 for Hard negative examples are hard, but useful

Figure 2 for Hard negative examples are hard, but useful

Figure 3 for Hard negative examples are hard, but useful

Figure 4 for Hard negative examples are hard, but useful

* CV, Triplet loss, Image embedding, 14 pages, 9 figures, ECCV 2020

Access Paper or Ask Questions

Visualizing How Embeddings Generalize

Sep 16, 2019
Xiaotong Liu, Hong Xuan, Zeyu Zhang, Abby Stylianou, Robert Pless

Figure 1 for Visualizing How Embeddings Generalize

Figure 2 for Visualizing How Embeddings Generalize

Figure 3 for Visualizing How Embeddings Generalize

Figure 4 for Visualizing How Embeddings Generalize

* 8 pages,4 figures, published in ICML workshop:Understanding and Improving Generalization in Deep Learning

Access Paper or Ask Questions

Deep Randomized Ensembles for Metric Learning

Sep 04, 2018
Hong Xuan, Richard Souvenir, Robert Pless

Figure 1 for Deep Randomized Ensembles for Metric Learning

Figure 2 for Deep Randomized Ensembles for Metric Learning

Figure 3 for Deep Randomized Ensembles for Metric Learning

Figure 4 for Deep Randomized Ensembles for Metric Learning

* ECCV 2018

Access Paper or Ask Questions