Memo: Test time robustness via adaptation and augmentation
While deep neural networks can attain good accuracy on in-distribution test points, many
applications require robustness even in the face of unexpected perturbations in the input …
applications require robustness even in the face of unexpected perturbations in the input …
The many faces of robustness: A critical analysis of out-of-distribution generalization
We introduce four new real-world distribution shift datasets consisting of changes in image
style, image blurriness, geographic location, camera operation, and more. With our new …
style, image blurriness, geographic location, camera operation, and more. With our new …
Improving robustness against common corruptions by covariate shift adaptation
Today's state-of-the-art machine vision models are vulnerable to image corruptions like
blurring or compression artefacts, limiting their performance in many real-world applications …
blurring or compression artefacts, limiting their performance in many real-world applications …
Self-training with noisy student improves imagenet classification
We present a simple self-training method that achieves 88.4% top-1 accuracy on ImageNet,
which is 2.0% better than the state-of-the-art model that requires 3.5 B weakly labeled …
which is 2.0% better than the state-of-the-art model that requires 3.5 B weakly labeled …
3d common corruptions and data augmentation
We introduce a set of image transformations that can be used as corruptions to evaluate the
robustness of models as well as data augmentation mechanisms for training neural …
robustness of models as well as data augmentation mechanisms for training neural …
Pretrained transformers improve out-of-distribution robustness
Although pretrained Transformers such as BERT achieve high accuracy on in-distribution
examples, do they generalize to new distributions? We systematically measure out-of …
examples, do they generalize to new distributions? We systematically measure out-of …
Self-supervised learning through the eyes of a child
Within months of birth, children develop meaningful expectations about the world around
them. How much of this early knowledge can be explained through generic learning …
them. How much of this early knowledge can be explained through generic learning …
On interaction between augmentations and corruptions in natural corruption robustness
Invariance to a broad array of image corruptions, such as warping, noise, or color shifts, is
an important aspect of building robust models in computer vision. Recently, several new …
an important aspect of building robust models in computer vision. Recently, several new …
Testing robustness against unforeseen adversaries
Most existing defenses against adversarial attacks only consider robustness to L_p-
bounded distortions. In reality, the specific attack is rarely known in advance and …
bounded distortions. In reality, the specific attack is rarely known in advance and …
Out-of-scope intent detection with self-supervision and discriminative training
Out-of-scope intent detection is of practical importance in task-oriented dialogue systems.
Since the distribution of outlier utterances is arbitrary and unknown in the training stage …
Since the distribution of outlier utterances is arbitrary and unknown in the training stage …