R
Richong Zhang
Researcher at Beihang University
Publications - 120
Citations - 2064
Richong Zhang is an academic researcher from Beihang University. The author has contributed to research in topics: Helpfulness & Knowledge base. The author has an hindex of 19, co-authored 118 publications receiving 1299 citations. Previous affiliations of Richong Zhang include Dalhousie University & University of Ottawa.
Papers
More filters
Journal ArticleDOI
MixUp as Locally Linear Out-Of-Manifold Regularization
TL;DR: In this paper, Zhang et al. developed an understanding for MixUp as a form of out-of-manifold regularization, which imposes certain linearity constraints on the model's input space beyond the data manifold.
Proceedings ArticleDOI
Aspect-Level Sentiment Analysis Via Convolution over Dependency Tree
TL;DR: A convolution over a dependency tree (CDT) model which exploits a Bi-directional Long Short Term Memory (Bi-LSTM) to learn representations for features of a sentence, and further enhance the embeddings with a graph convolutional network (GCN) which operates directly on the dependency tree of the sentence.
Posted Content
Augmenting Data with Mixup for Sentence Classification: An Empirical Study
TL;DR: Two strategies for the adaption of Mixup on sentence classification are proposed: one performs interpolation on word embeddings and another on sentence embedDings, and both serve as an effective, domain independent data augmentation approach for sentence classification.
Proceedings ArticleDOI
A Novel Approach for API Recommendation in Mashup Development
TL;DR: A probabilistic model to assist mashup creators by recommending a list of APIs that may be used to compose a required mashup given descriptions of the mashup is proposed.
Posted Content
MixUp as Locally Linear Out-Of-Manifold Regularization
TL;DR: An understanding is developed for MixUp as a form of “out-of-manifold regularization”, which imposes certain “local linearity” constraints on the model’s input space beyond the data manifold, which enables a novel adaptive version of MixUp, where the mixing policies are automatically learned from the data using an additional network and objective function designed to avoid manifold intrusion.