Boosting Adversarial Attacks with Momentum
- Yinpeng Dong, Fangzhou Liao, Jianguo Li
- Computer ScienceIEEE/CVF Conference on Computer Vision and…
- 17 October 2017
A broad class of momentum-based iterative algorithms to boost adversarial attacks by integrating the momentum term into the iterative process for attacks, which can stabilize update directions and escape from poor local maxima during the iterations, resulting in more transferable adversarial examples.
Adversarial Attack on Graph Structured Data
This paper proposes a reinforcement learning based attack method that learns the generalizable attack policy, while only requiring prediction labels from the target classifier, and uses both synthetic and real-world data to show that a family of Graph Neural Network models are vulnerable to adversarial attacks.
Evading Defenses to Transferable Adversarial Examples by Translation-Invariant Attacks
- Yinpeng Dong, Tianyu Pang, Hang Su, Jun Zhu
- Computer ScienceComputer Vision and Pattern Recognition
- 5 April 2019
A translation-invariant attack method to generate more transferable adversarial examples against the defense models, which fools eight state-of-the-art defenses at an 82% success rate on average based only on the transferability, demonstrating the insecurity of the current defense techniques.
Defense Against Adversarial Attacks Using High-Level Representation Guided Denoiser
- Fangzhou Liao, Ming Liang, Yinpeng Dong, Tianyu Pang, Jun Zhu, Xiaolin Hu
- Computer ScienceIEEE/CVF Conference on Computer Vision and…
- 8 December 2017
High-level representation guided denoiser (HGD) is proposed as a defense for image classification by using a loss function defined as the difference between the target model's outputs activated by the clean image and denoised image.
Stochastic Training of Graph Convolutional Networks with Variance Reduction
- Jianfei Chen, Jun Zhu, Le Song
- Computer ScienceInternational Conference on Machine Learning
- 29 October 2017
Control variate based algorithms which allow sampling an arbitrarily small neighbor size are developed and a new theoretical guarantee for these algorithms to converge to a local optimum of GCN is proved.
Triple Generative Adversarial Nets
- Chongxuan Li, T. Xu, Jun Zhu, Bo Zhang
- Computer ScienceNIPS
- 7 March 2017
Triple-GAN as a unified model can simultaneously achieve the state-of-the-art classification results among deep generative models, and disentangle the classes and styles of the input and transfer smoothly in the data space via interpolation in the latent space class-conditionally.
Improving Adversarial Robustness via Promoting Ensemble Diversity
- Tianyu Pang, Kun Xu, Chao Du, Ning Chen, Jun Zhu
- Computer ScienceInternational Conference on Machine Learning
- 25 January 2019
A new notion of ensemble diversity in the adversarial setting is defined as the diversity among non-maximal predictions of individual members, and an adaptive diversity promoting (ADP) regularizer is presented to encourage the diversity, which leads to globally better robustness for the ensemble by making adversarial examples difficult to transfer among individual members.
Towards Better Analysis of Deep Convolutional Neural Networks
- Mengchen Liu, Jiaxin Shi, Z. Li, Chongxuan Li, Jun Zhu, Shixia Liu
- Computer ScienceIEEE Transactions on Visualization and Computer…
- 24 April 2016
A hybrid visualization is developed to disclose the multiple facets of each neuron and the interactions between them and a biclustering-based edge bundling method is proposed to reduce visual clutter caused by a large number of connections between neurons.
Sparse Topical Coding
The results demonstrate the advantages of STC and supervised MedSTC on identifying topical meanings of words and improving classification accuracy and time efficiency.
MedLDA: maximum margin supervised topic models
The maximum entropy discrimination latent Dirichlet allocation (MedLDA) model is proposed, which integrates the mechanismbehind the max-margin prediction models with the mechanism behind the hierarchical Bayesian topic models under a unified constrained optimization framework, and yields latent topical representations that are more discriminative and more suitable for prediction tasks such as document classification or regression.
...
...