activity
20062023
most citedAdversarial Attack and Defense on Graph Data: A Survey

293 citations · 1.5k across the 77 of their papers we have counts for

collaborators
Showing 2020Show all

8 papers · 1 filter

astro-ph.SR2020★ 10 cited

Mixed properties of magnetohydrodynamic waves undergoing resonant absorption in the cusp continuum

Marcel Goossens, Shao-Xia Chen, Michaël Geeraerts +2

Observations of magnetohydrodynamic (MHD) waves in the structured solar atmosphere have shown that these waves are damped and can thus contribute to atmospheric heating. In this pa…

astro-ph.SR2020★ 15 cited

How eruptions of a small filament feed materials to a nearby larger-scaled filament

Hengyuan Wei, Zhenghua Huang, Zhenyong Hou +4

As one of the most common features in the solar atmosphere, filaments are significant not only in the solar physics but also in the stellar and laboratory plasma physics. With the…

cs.CL2020

Adversarial Mutual Information for Text Generation

Boyuan Pan, Yazheng Yang, Kaizhao Liang +5

Recent advances in maximizing mutual information (MI) between the source and target have demonstrated its effectiveness in text generation. However, previous works paid little atte…

stat.ML2020

Weighted Lasso Estimates for Sparse Logistic Regression: Non-asymptotic Properties with Measurement Error

Huamei Huang, Yujing Gao, Huiming Zhang +1

When we are interested in high-dimensional system and focus on classification performance, the -penalized logistic regression is becoming important and popular. However,…

cs.LG2020★ 3 cited

Stable Prediction via Leveraging Seed Variable

Kun Kuang, Bo Li, Peng Cui +4

In this paper, we focus on the problem of stable prediction across unknown test data, where the test distribution is agnostic and might be totally different from the training one.…

cs.LG2020★ 2 cited

Uncovering the Connections Between Adversarial Transferability and Knowledge Transferability

Kaizhao Liang, Jacky Y. Zhang, Boxin Wang +3

Knowledge transferability, or transfer learning, has been widely adopted to allow a pre-trained model in the source domain to be effectively adapted to downstream tasks in the targ…