Polarizing Front Ends for Robust CNNs

• URL: http://arxiv.org/abs/2002.09580v1
• Date: Sat, 22 Feb 2020 00:28:41 GMT
• Title: Polarizing Front Ends for Robust CNNs
• Authors: Can Bakiskan, Soorya Gopalakrishnan, Metehan Cekic, Upamanyu Madhow, Ramtin Pedarsani
• Abstract summary: We propose a bottom-up strategy for attenuating adversarial perturbations using a nonlinear front end which polarizes and quantizes the data. We observe that ideal polarization can be utilized to completely eliminate perturbations, develop algorithms to learn approximately polarizing bases for data, and investigate the effectiveness of the proposed strategy on the MNIST and Fashion MNIST datasets.
• Score: 23.451381552751393
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: The vulnerability of deep neural networks to small, adversarially designed perturbations can be attributed to their "excessive linearity." In this paper, we propose a bottom-up strategy for attenuating adversarial perturbations using a nonlinear front end which polarizes and quantizes the data. We observe that ideal polarization can be utilized to completely eliminate perturbations, develop algorithms to learn approximately polarizing bases for data, and investigate the effectiveness of the proposed strategy on the MNIST and Fashion MNIST datasets.

Related papers

• Deep Polarization Cues for Single-shot Shape and Subsurface Scattering Estimation [13.561603248769302]
  We propose a novel learning-based method to jointly estimate the shape and subsurface scattering (SSS) parameters of translucent objects. Our observations indicate that the SSS affects not only the light intensity but also the polarization signal. We introduce the first large-scale synthetic dataset of polarized translucent objects for training our model.
  arXiv  Detail & Related papers  (2024-07-11T03:00:24Z)
• Robust Depth Enhancement via Polarization Prompt Fusion Tuning [112.88371907047396]
  We present a framework that leverages polarization imaging to improve inaccurate depth measurements from various depth sensors. Our method first adopts a learning-based strategy where a neural network is trained to estimate a dense and complete depth map from polarization data and a sensor depth map from different sensors. To further improve the performance, we propose a Polarization Prompt Fusion Tuning (PPFT) strategy to effectively utilize RGB-based models pre-trained on large-scale datasets.
  arXiv  Detail & Related papers  (2024-04-05T17:55:33Z)
• Mitigating Framing Bias with Polarity Minimization Loss [56.24404488440295]
  Framing bias plays a significant role in exacerbating political polarization by distorting the perception of actual events. We propose a new loss function that encourages the model to minimize the polarity difference between the polarized input articles to reduce framing bias.
  arXiv  Detail & Related papers  (2023-11-03T09:50:23Z)
• PARTNER: Level up the Polar Representation for LiDAR 3D Object Detection [81.16859686137435]
  We present PARTNER, a novel 3D object detector in the polar coordinate. Our method outperforms the previous polar-based works with remarkable margins of 3.68% and 9.15% on and ONCE validation set.
  arXiv  Detail & Related papers  (2023-08-08T01:59:20Z)
• Neural Polarizer: A Lightweight and Effective Backdoor Defense via Purifying Poisoned Features [62.82817831278743]
  Recent studies have demonstrated the susceptibility of deep neural networks to backdoor attacks. We propose a novel backdoor defense method by inserting a learnable neural polarizer into the backdoored model as an intermediate layer.
  arXiv  Detail & Related papers  (2023-06-29T05:39:58Z)
• Physically-admissible polarimetric data augmentation for road-scene analysis [4.972086627584208]
  We propose CycleGAN to transfer large labeled road scene datasets to the polarimetric domain. The resulting constrained CycleGAN is publicly released, allowing anyone to generate their own polarimetric images.
  arXiv  Detail & Related papers  (2022-06-15T10:04:43Z)
• Polarized skylight orientation determination artificial neural network [4.834173456342489]
  This paper proposes an artificial neural network to determine orientation using polarized skylight. The degree of polarization (DOP) and angle of polarization (AOP) are directly extracted in the network.
  arXiv  Detail & Related papers  (2021-07-06T00:19:22Z)
• Attribute-Guided Adversarial Training for Robustness to Natural Perturbations [64.35805267250682]
  We propose an adversarial training approach which learns to generate new samples so as to maximize exposure of the classifier to the attributes-space. Our approach enables deep neural networks to be robust against a wide range of naturally occurring perturbations.
  arXiv  Detail & Related papers  (2020-12-03T10:17:30Z)
• Optimizing Mode Connectivity via Neuron Alignment [84.26606622400423]
  Empirically, the local minima of loss functions can be connected by a learned curve in model space along which the loss remains nearly constant. We propose a more general framework to investigate effect of symmetry on landscape connectivity by accounting for the weight permutations of networks being connected.
  arXiv  Detail & Related papers  (2020-09-05T02:25:23Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.