Hanze Dong

Hanze Dong
The Hong Kong University of Science and Technology | UST · Department of Mathematics

Bachelor of Science

About

10
Publications
706
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
25
Citations
Additional affiliations
January 2019 - July 2023
The Hong Kong University of Science and Technology
Position
  • PhD Student

Publications

Publications (10)
Article
This paper studies the problem of Generalized Zero-shot Learning (G-ZSL), whose goal is to classify instances from both seen and unseen classes at the test time. We propose a novel domain division method to solve G-ZSL. Some previous models with domain division operations only calibrate the confident prediction of source classes (W-SVM (Scheirer et...
Preprint
The communication of gradients is a key bottleneck in distributed training of large scale machine learning models. In order to reduce the communication cost, gradient compression (e.g., sparsification and quantization) and error compensation techniques are often used. In this paper, we propose and study three new efficient methods in this space: er...
Article
Full-text available
Deep learning has received considerable empirical success in recent years. However, while many ad hoc tricks have been discovered by practitioners, until recently, there has been a lack of theoretical understanding for tricks invented in the deep learning literature. Known by practitioners that overparameterized neural networks (NNs) are easy to le...
Preprint
Deep learning has received considerable empirical successes in recent years. However, while many ad hoc tricks have been discovered by practitioners, until recently, there has been a lack of theoretical understanding for tricks invented in the deep learning literature. Known by practitioners that overparameterized neural networks are easy to learn,...
Preprint
This paper proposes a Disentangled gEnerative cAusal Representation (DEAR) learning method. Unlike existing disentanglement methods that enforce independence of the latent variables, we consider the general case where the underlying factors of interests can be causally correlated. We show that previous methods with independent priors fail to disent...
Article
Regarding extreme value theory, the unseen novel classes in the open-set recognition can be seen as the extreme values of training classes. Following this idea, we introduce the margin and coverage distribution to model the training classes. A novel visual-semantic embedding framework — extreme vocabulary learning (EVoL) is proposed; the EVoL embed...
Preprint
Recently, over-parameterized neural networks have been extensively analyzed in the literature. However, the previous studies cannot satisfactorily explain why fully trained neural networks are successful in practice. In this paper, we present a new theoretical framework for analyzing over-parameterized neural networks which we call neural feature r...
Article
Despite significant progress in object categorization, in recent years, a number of important challenges remain; mainly, ability to learn from limited labeled data and ability to recognize object classes within large, potentially open, set of labels. Zero-shot learning is one way of addressing these challenges, but it has only been shown to work wi...
Preprint
This paper studies the problem of domain division problem which aims to segment instances drawn from different probabilistic distributions. Such a problem exists in many previous recognition tasks, such as Open Set Learning (OSL) and Generalized Zero-Shot Learning (G-ZSL), where the testing instances come from either seen or novel/unseen classes of...
Article
The novel unseen classes can be formulated as the extreme values of known classes. This inspired the recent works on open-set recognition \cite{Scheirer_2013_TPAMI,Scheirer_2014_TPAMIb,EVM}, which however can have no way of naming the novel unseen classes. To solve this problem, we propose the Extreme Value Learning (EVL) formulation to learn the m...