Alert button
Picture for Andi Zhang

Andi Zhang

Alert button

Constructing Semantics-Aware Adversarial Examples with Probabilistic Perspective

Jun 01, 2023
Andi Zhang, Damon Wischik

Figure 1 for Constructing Semantics-Aware Adversarial Examples with Probabilistic Perspective
Figure 2 for Constructing Semantics-Aware Adversarial Examples with Probabilistic Perspective
Figure 3 for Constructing Semantics-Aware Adversarial Examples with Probabilistic Perspective
Figure 4 for Constructing Semantics-Aware Adversarial Examples with Probabilistic Perspective

In this study, we introduce a novel, probabilistic viewpoint on adversarial examples, achieved through box-constrained Langevin Monte Carlo (LMC). Proceeding from this perspective, we develop an innovative approach for generating semantics-aware adversarial examples in a principled manner. This methodology transcends the restriction imposed by geometric distance, instead opting for semantic constraints. Our approach empowers individuals to incorporate their personal comprehension of semantics into the model. Through human evaluation, we validate that our semantics-aware adversarial examples maintain their inherent meaning. Experimental findings on the MNIST and SVHN datasets demonstrate that our semantics-aware adversarial examples can effectively circumvent robust adversarial training methods tailored for traditional adversarial attacks.

* 17 pages, 14 figures 
Viaarxiv icon

SR-OOD: Out-of-Distribution Detection via Sample Repairing

May 26, 2023
Rui Sun, Andi Zhang, Haiming Zhang, Yao Zhu, Ruimao Zhang, Zhen Li

Figure 1 for SR-OOD: Out-of-Distribution Detection via Sample Repairing
Figure 2 for SR-OOD: Out-of-Distribution Detection via Sample Repairing
Figure 3 for SR-OOD: Out-of-Distribution Detection via Sample Repairing
Figure 4 for SR-OOD: Out-of-Distribution Detection via Sample Repairing

It is widely reported that deep generative models can classify out-of-distribution (OOD) samples as in-distribution with high confidence. In this work, we propose a hypothesis that this phenomenon is due to the reconstruction task, which can cause the generative model to focus too much on low-level features and not enough on semantic information. To address this issue, we introduce SR-OOD, an OOD detection framework that utilizes sample repairing to encourage the generative model to learn more than just an identity map. By focusing on semantics, our framework improves OOD detection performance without external data and label information. Our experimental results demonstrate the competitiveness of our approach in detecting OOD samples.

Viaarxiv icon

Falsehoods that ML researchers believe about OOD detection

Nov 01, 2022
Andi Zhang, Damon Wischik

An intuitive way to detect out-of-distribution (OOD) data is via the density function of a fitted probabilistic generative model: points with low density may be classed as OOD. But this approach has been found to fail, in deep learning settings. In this paper, we list some falsehoods that machine learning researchers believe about density-based OOD detection. Many recent works have proposed likelihood-ratio-based methods to `fix' the problem. We propose a framework, the OOD proxy framework, to unify these methods, and we argue that likelihood ratio is a principled method for OOD detection and not a mere `fix'. Finally, we discuss the relationship between domain discrimination and semantics.

* 5 pages 
Viaarxiv icon

Out-of-Distribution Detection with Class Ratio Estimation

Jun 08, 2022
Mingtian Zhang, Andi Zhang, Tim Z. Xiao, Yitong Sun, Steven McDonagh

Figure 1 for Out-of-Distribution Detection with Class Ratio Estimation
Figure 2 for Out-of-Distribution Detection with Class Ratio Estimation
Figure 3 for Out-of-Distribution Detection with Class Ratio Estimation
Figure 4 for Out-of-Distribution Detection with Class Ratio Estimation

Density-based Out-of-distribution (OOD) detection has recently been shown unreliable for the task of detecting OOD images. Various density ratio based approaches achieve good empirical performance, however methods typically lack a principled probabilistic modelling explanation. In this work, we propose to unify density ratio based methods under a novel framework that builds energy-based models and employs differing base distributions. Under our framework, the density ratio can be viewed as the unnormalized density of an implicit semantic distribution. Further, we propose to directly estimate the density ratio of a data sample through class ratio estimation. We report competitive results on OOD image problems in comparison with recent work that alternatively requires training of deep generative models for the task. Our approach enables a simple and yet effective path towards solving the OOD detection problem.

Viaarxiv icon

A Dual-fusion Semantic Segmentation Framework With GAN For SAR Images

Jun 02, 2022
Donghui Li, Jia Liu, Fang Liu, Wenhua Zhang, Andi Zhang, Wenfei Gao, Jiao Shi

Figure 1 for A Dual-fusion Semantic Segmentation Framework With GAN For SAR Images
Figure 2 for A Dual-fusion Semantic Segmentation Framework With GAN For SAR Images
Figure 3 for A Dual-fusion Semantic Segmentation Framework With GAN For SAR Images
Figure 4 for A Dual-fusion Semantic Segmentation Framework With GAN For SAR Images

Deep learning based semantic segmentation is one of the popular methods in remote sensing image segmentation. In this paper, a network based on the widely used encoderdecoder architecture is proposed to accomplish the synthetic aperture radar (SAR) images segmentation. With the better representation capability of optical images, we propose to enrich SAR images with generated optical images via the generative adversative network (GAN) trained by numerous SAR and optical images. These optical images can be used as expansions of original SAR images, thus ensuring robust result of segmentation. Then the optical images generated by the GAN are stitched together with the corresponding real images. An attention module following the stitched data is used to strengthen the representation of the objects. Experiments indicate that our method is efficient compared to other commonly used methods

* 4 pages,4 figures, 2022 IEEE International Geoscience and Remote Sensing Symposium 
Viaarxiv icon

Measuring Uncertainty in Signal Fingerprinting with Gaussian Processes Going Deep

Sep 10, 2021
Ran Guan, Andi Zhang, Mengchao Li, Yongliang Wang

Figure 1 for Measuring Uncertainty in Signal Fingerprinting with Gaussian Processes Going Deep
Figure 2 for Measuring Uncertainty in Signal Fingerprinting with Gaussian Processes Going Deep
Figure 3 for Measuring Uncertainty in Signal Fingerprinting with Gaussian Processes Going Deep
Figure 4 for Measuring Uncertainty in Signal Fingerprinting with Gaussian Processes Going Deep

In indoor positioning, signal fluctuation is highly location-dependent. However, signal uncertainty is one critical yet commonly overlooked dimension of the radio signal to be fingerprinted. This paper reviews the commonly used Gaussian Processes (GP) for probabilistic positioning and points out the pitfall of using GP to model signal fingerprint uncertainty. This paper also proposes Deep Gaussian Processes (DGP) as a more informative alternative to address the issue. How DGP better measures uncertainty in signal fingerprinting is evaluated via simulated and realistically collected datasets.

* 8 pages, 10 figures; To be appear on the 2021 International Conference on Indoor Positioning and Indoor Navigation (IPIN) 
Viaarxiv icon

On the Out-of-distribution Generalization of Probabilistic Image Modelling

Sep 04, 2021
Mingtian Zhang, Andi Zhang, Steven McDonagh

Figure 1 for On the Out-of-distribution Generalization of Probabilistic Image Modelling
Figure 2 for On the Out-of-distribution Generalization of Probabilistic Image Modelling
Figure 3 for On the Out-of-distribution Generalization of Probabilistic Image Modelling
Figure 4 for On the Out-of-distribution Generalization of Probabilistic Image Modelling

Out-of-distribution (OOD) detection and lossless compression constitute two problems that can be solved by the training of probabilistic models on a first dataset with subsequent likelihood evaluation on a second dataset, where data distributions differ. By defining the generalization of probabilistic models in terms of likelihood we show that, in the case of image models, the OOD generalization ability is dominated by local features. This motivates our proposal of a Local Autoregressive model that exclusively models local image features towards improving OOD performance. We apply the proposed model to OOD detection tasks and achieve state-of-the-art unsupervised OOD detection performance without the introduction of additional data. Additionally, we employ our model to build a new lossless image compressor: NeLLoC (Neural Local Lossless Compressor) and report state-of-the-art compression rates and model size.

Viaarxiv icon

Memory-augmented Neural Machine Translation

Aug 07, 2017
Yang Feng, Shiyue Zhang, Andi Zhang, Dong Wang, Andrew Abel

Figure 1 for Memory-augmented Neural Machine Translation
Figure 2 for Memory-augmented Neural Machine Translation
Figure 3 for Memory-augmented Neural Machine Translation
Figure 4 for Memory-augmented Neural Machine Translation

Neural machine translation (NMT) has achieved notable success in recent times, however it is also widely recognized that this approach has limitations with handling infrequent words and word pairs. This paper presents a novel memory-augmented NMT (M-NMT) architecture, which stores knowledge about how words (usually infrequently encountered ones) should be translated in a memory and then utilizes them to assist the neural model. We use this memory mechanism to combine the knowledge learned from a conventional statistical machine translation system and the rules learned by an NMT system, and also propose a solution for out-of-vocabulary (OOV) words based on this framework. Our experiments on two Chinese-English translation tasks demonstrated that the M-NMT architecture outperformed the NMT baseline by $9.0$ and $2.7$ BLEU points on the two tasks, respectively. Additionally, we found this architecture resulted in a much more effective OOV treatment compared to competitive methods.

Viaarxiv icon

Flexible and Creative Chinese Poetry Generation Using Neural Memory

May 10, 2017
Jiyuan Zhang, Yang Feng, Dong Wang, Yang Wang, Andrew Abel, Shiyue Zhang, Andi Zhang

Figure 1 for Flexible and Creative Chinese Poetry Generation Using Neural Memory
Figure 2 for Flexible and Creative Chinese Poetry Generation Using Neural Memory
Figure 3 for Flexible and Creative Chinese Poetry Generation Using Neural Memory
Figure 4 for Flexible and Creative Chinese Poetry Generation Using Neural Memory

It has been shown that Chinese poems can be successfully generated by sequence-to-sequence neural models, particularly with the attention mechanism. A potential problem of this approach, however, is that neural models can only learn abstract rules, while poem generation is a highly creative process that involves not only rules but also innovations for which pure statistical models are not appropriate in principle. This work proposes a memory-augmented neural model for Chinese poem generation, where the neural model and the augmented memory work together to balance the requirements of linguistic accordance and aesthetic innovation, leading to innovative generations that are still rule-compliant. In addition, it is found that the memory mechanism provides interesting flexibility that can be used to generate poems with different styles.

Viaarxiv icon