paper-with-me

홈 › Papers

Why are Saliency Maps Noisy? Cause of and Solution to Noisy Saliency Maps

2019-02-13 · Beomsu Kim, Junghoon Seo, SeungHyun Jeon, Jamyoung Koo, Jeongyeol Choe, Taegyun Jeon

Saliency Map, the gradient of the score function with respect to the input, is the most basic technique for interpreting deep neural network decisions. However, saliency maps are often visually noisy. Although several hypotheses were proposed to account for this phenomenon, there are few works that provide rigorous analyses of noisy saliency maps. In this paper, we firstly propose a new hypothesis that noise may occur in saliency maps when irrelevant features pass through ReLU activation functions. Then, we propose Rectified Gradient, a method that alleviates this problem through layer-wise thresholding during backpropagation. Experiments with neural networks trained on CIFAR-10 and ImageNet showed effectiveness of our method and its superiority to other attribution methods.

📄 PDF Abstract BibTeX arXiv:1902.04893

Code (2)

1202kbs/Rectified-Gradient 공식 구현 tf
lenbrocki/NoBias-Rectified-Gradient tf

Methods 이 논문이 사용한 방법론

ReLU How Do I Communicate to Expedia? How Do I Communicate to Expedia? – Call ☎️ +1-(888) 829 (0881) or +1-805-330-4056 or +1-805-330-4056 for Live Support & Special Travel…

Similar Papers 제목 키워드 기반

Rectified Gradient: Layer-wise Thresholding for Sharp and Coherent Attribution Maps

2018-09-27 · Beomsu Kim, Junghoon Seo, Jeongyeol Choe, Jamyoung Koo 외

Saliency map, or the gradient of the score function with respect to the input, is the most basic means of interpreting deep neural network decisions. However, saliency maps are often visually noisy. Although several hypo…

Learning Noise-Aware Encoder-Decoder from Noisy Labels by Alternating Back-Propagation for Saliency Detection

2020-07-23 · ECCV 2020 8 · Jing Zhang, Jianwen Xie, Nick Barnes

In this paper, we propose a noise-aware encoder-decoder framework to disentangle a clean saliency predictor from noisy training examples, where the noisy labels are generated by unsupervised handcrafted feature-based met…

DecoderSaliency Detection

Deep Unsupervised Saliency Detection: A Multiple Noisy Labeling Perspective

2018-03-29 · CVPR 2018 6 · Jing Zhang, Tong Zhang, Yuchao Dai, Mehrtash Harandi 외

The success of current deep saliency detection methods heavily depends on the availability of large-scale supervision in the form of per-pixel labeling. Such supervision, while labor-intensive and not always possible, te…

BenchmarkingSaliency DetectionSaliency PredictionUnsupervised Saliency Detection

DSAL-GAN: Denoising based Saliency Prediction with Generative Adversarial Networks

2019-04-02 · Prerana Mukherjee, Manoj Sharma, Megh Makwana, Ajay Pratap Singh 외

Synthesizing high quality saliency maps from noisy images is a challenging problem in computer vision and has many practical applications. Samples generated by existing techniques for saliency detection cannot handle the…

DenoisingGenerative Adversarial Networkobject-detectionObject Detection+5

Improving Deep Learning Interpretability by Saliency Guided Training

2021-11-29 · NeurIPS 2021 12 · Aya Abdelsalam Ismail, Héctor Corrada Bravo, Soheil Feizi

Saliency methods have been widely used to highlight important input features in model predictions. Most existing methods use backpropagation on a modified gradient function to generate saliency maps. Thus, noisy gradient…

Deep LearningTime SeriesTime Series Analysis