Research map / Adversarial robustness and security / Attacks on text and graphs
Learning from brains how to regularize machines
Zhe Li, Wieland Brendel, Edgar Walker, Erick Cobos, Taliah Muhammad, Jacob Reimer, Matthias Bethge, Fabian Sinz, Zachary Pitkow, Andreas Tolias
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add neural similarity regularization (from predictive model) | ResNet18 · CIFAR10 Gaussian noise, highest noise level | accuracy: ~20% -> ~50% | Improved |
| add neural similarity regularization | ResNet18 · CIFAR10 adversarial L2 | median perturbation: 0.09 -> 0.13 | Improved |
| add neural similarity regularization | ResNet18 · CIFAR10 adversarial Linf | median perturbation: 0.0025 -> 0.0034 | Improved |
| use shuffled neural similarity matrix instead of neural one (ablation) | Neural regularization on ResNet18 · CIFAR10 Gaussian noise | accuracy | Hurt |
| use pretrained VGG19 conv3-1 similarity instead of neural one (ablation) | Neural regularization on ResNet18 · CIFAR10 adversarial Linf | median perturbation: 0.0034 -> 0.0028 | Hurt |
| use raw neural data similarity instead of predictive-model denoised similarity (ablation) | Neural regularization on ResNet18 · CIFAR10 Gaussian noise | accuracy | Hurt |
| add neural regularization | ResNet34 · grayscale CIFAR100 Gaussian noise | accuracy | Improved |
| increase regularization strength alpha (ablation) | Neural regularization on ResNet18 · CIFAR10 Gaussian noise | accuracy | Improved |
| add neural regularization | ResNet18 · CIFAR10 clean images | accuracy | No clear effect |
Closest papers by meaning
- Towards robust vision by multi-task learning on monkey visual cortex (NeurIPS 2021 · similarity 0.89)
- Reproducibility of predictive networks for mouse visual cortex (NeurIPS 2024 · similarity 0.89)
- Understanding Robust Learning through the Lens of Representation Similarities (NeurIPS 2022 · similarity 0.88)
- Neural Regression, Representational Similarity, Model Zoology & Neural Taskonomy at Scale in Rodent Visual Cortex (NeurIPS 2021 · similarity 0.88)
- Improving the Adversarial Robustness and Interpretability of Deep Neural Networks by Regularizing Their Input Gradients (AAAI 2018 · similarity 0.87)
- Comparison Against Task Driven Artificial Neural Networks Reveals Functional Properties in Mouse Visual Cortex (NeurIPS 2019 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
