A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.
Diffusion models outperform GANs in image synthesis quality.
We study the robustness of image classifiers to temporal perturbations derived from videos. As part of this study, we construct two datasets, ImageNet-Vid-Robust and YTBB-Robust , containing a total 57,897 images grouped into 3,139 sets of perceptually similar images. Our datasets were derived from ImageNet-Vid and You…
In this paper we establish rigorous benchmarks for image classifier robustness. Our first benchmark, ImageNet-C, standardizes and expands the corruption robustness topic, while showing which classifiers are preferable in safety-critical applications. Then we propose a new dataset called ImageNet-P which enables researc…
Work on making classifiers robust against adversarial attacks for top-k predictions.
In this work we develop a fast saliency detection method that can be applied to any differentiable image classifier. We train a masking model to manipulate the scores of the classifier by masking salient parts of the input image. Our model generalises well to unseen images and requires a single forward pass to perform …
We build new test sets for the CIFAR-10 and ImageNet datasets. Both benchmarks have been the focus of intense research for almost a decade, raising the danger of overfitting to excessively re-used test sets. By closely following the original dataset creation processes, we test to what extent current classification mode…
We show how to turn any classifier that classifies well under Gaussian noise into a new classifier that is certifiably robust to adversarial perturbations under the norm. This "randomized smoothing" technique has been proposed recently in the literature, but existing guarantees are loose. We prove a tight robu…
New method improves robustness of smoothed classifiers.
New method provides formal uncertainty guarantees for image classifiers.
MALT improves adversarial attacks by targeting classes more efficiently.
LEEP measures transferability of learned representations efficiently.
New method to evaluate visual explanations from neural networks.
New bounds show robust models can generalize well, contrary to prior theories.
New method improves certified robustness for classifier confidence.
Meta-Cal improves post-hoc calibration of neural networks.
Deep Neural Networks (DNNs) excel on many complex perceptual tasks but it has proven notoriously difficult to understand how they reach their decisions. We here introduce a high-performance DNN architecture on ImageNet whose decisions are considerably easier to explain. Our model, a simple variant of the ResNet-50 arch…
New bound proves rationality helps generalization in self-supervised learning.
New samplers improve compositional generation with diffusion models.
CNNs overinterpret inputs, leading to high accuracy without meaningful features.
ViewFool identifies adversarial viewpoints to test image recognition robustness.
We introduce a sparse scattering deep convolutional neural network, which provides a simple model to analyze properties of deep representation learning for classification. Learning a single dictionary matrix with a classifier yields a higher classification accuracy than AlexNet over the ImageNet 2012 dataset. The netwo…
Recent works have shown the effectiveness of randomized smoothing as a scalable technique for building neural network-based classifiers that are provably robust to -norm adversarial perturbations. In this paper, we employ adversarial training to improve the performance of randomized smoothing. We design an adap…
Metric-based few-shot learning methods try to overcome the difficulty due to the lack of training examples by learning embedding to make comparison easy. We propose a novel algorithm to generate class representatives for few-shot classification tasks. As a probabilistic model for learned features of inputs, we consider…
Synthesizing high resolution photorealistic images has been a long-standing challenge in machine learning. In this paper we introduce new methods for the improved training of generative adversarial networks (GANs) for image synthesis. We construct a variant of GANs employing label conditioning that results in 128x128 r…
Paper introduces SPADE method to protect classifiers from OOD and adversarial samples.
Perturbation-based explanation methods often measure the contribution of an input feature to an image classifier's outputs by heuristically removing it via e.g. blurring, adding noise, or graying out, which often produce unrealistic, out-of-samples. Instead, we propose to integrate a generative inpainter into three rep…
Deep AD on images outperforms traditional methods.
Adversarial perturbations dramatically decrease the accuracy of state-of-the-art image classifiers. In this paper, we propose and analyze a simple and computationally efficient defense strategy: inject random Gaussian noise, discretize each pixel, and then feed the result into any pre-trained classifier. Theoretically,…
While deep neural networks have proven to be a powerful tool for many recognition and classification tasks, their stability properties are still not well understood. In the past, image classifiers have been shown to be vulnerable to so-called adversarial attacks, which are created by additively perturbing the correctly…
Denoised smoothing defends pretrained classifiers against adversarial attacks.
Neural nets learn simple distributions first, then more complex ones.
In this paper we establish rigorous benchmarks for image classifier robustness. Our first benchmark, ImageNet-C, standardizes and expands the corruption robustness topic, while showing which classifiers are preferable in safety-critical applications. Unlike recent robustness research, this benchmark evaluates performan…
Current neural network-based classifiers are susceptible to adversarial examples even in the black-box setting, where the attacker only has query access to the model. In practice, the threat model for real-world systems is often more restrictive than the typical black-box model where the adversary can observe the full …
Currently available methods for extracting saliency maps identify parts of the input which are the most important to a specific fixed classifier. We show that this strong dependence on a given classifier hinders their performance. To address this problem, we propose classifier-agnostic saliency map extraction, which fi…
EBM life cycle project improves MCMC for image generation, defense, and density modeling.
Improved image classification accuracy with a probabilistic model of label noise.
Improves GANs by incorporating class information with a multi-hinge loss.
HET-XL improves heteroscedastic classifiers for large-scale image classification.
We investigate the robustness properties of ResNeXt class image recognition models trained with billion scale weakly supervised data (ResNeXt WSL models). These models, recently made public by Facebook AI, were trained with ~1B images from Instagram and fine-tuned on ImageNet. We show that these models display an unpre…
MCAL reduces labeling costs by 6x for auto-labeling data sets.
Unified Uncertainty Calibration improves AI predictions by combining different types of uncertainty.
We introduce collaborative learning in which multiple classifier heads of the same network are simultaneously trained on the same training data to improve generalization and robustness to label noise with no extra inference cost. It acquires the strengths from auxiliary training, multi-task learning and knowledge disti…
Data augmentation is an effective technique for improving the accuracy of modern image classifiers. However, current data augmentation implementations are manually designed. In this paper, we describe a simple procedure called AutoAugment to automatically search for improved data augmentation policies. In our implement…
An unsupervised learning classification model is described. It achieves classification error probability competitive with that of popular supervised learning classifiers such as SVM or kNN. The model is based on the incremental execution of small step shift and rotation operations upon selected discriminative hyperplan…
Proposes HetSNGP method for joint model and data uncertainty modeling.
New method certifies robustness to sparse adversarial attacks.
Transfer learning is a cornerstone of computer vision, yet little work has been done to evaluate the relationship between architecture and transfer. An implicit hypothesis in modern computer vision research is that models that perform better on ImageNet necessarily perform better on other vision tasks. However, this hy…
We study the problem of learning classifiers robust to universal adversarial perturbations. While prior work approaches this problem via robust optimization, adversarial training, or input transformation, we instead phrase it as a two-player zero-sum game. In this new formulation, both players simultaneously play the s…