We often see undesirable tradeoffs in robust machine learning where out-of-distribution (OOD) accuracy is at odds with in-distribution (ID) accuracy: a robust classifier obtained via specialized techniques such as removing spurious features often has better OOD but worse ID accuracy compared to a standard classifier trained via ERM. In this paper, we find that ID-calibrated ensembles -- where we simply ensemble the standard and robust models after calibrating on only ID data -- outperforms prior state-of-the-art (based on self-training) on both ID and OOD accuracy. On eleven natural distribution shift datasets, ID-calibrated ensembles obtain the best of both worlds: strong ID accuracy and OOD accuracy. We analyze this method in stylized set...
Machine learning algorithms typically assume that training and test examples are drawn from the same...
Unsupervised out-of-distribution (U-OOD) detection has recently attracted much attention due its imp...
While deep learning models have seen widespread success in controlled environments, there are still ...
The ability to detect Out-of-Distribution (OOD) data is important in safety-critical applications of...
In the problem of out-of-distribution (OOD) detection, the usage of auxiliary data as outlier data f...
Existing works have made great progress in improving adversarial robustness, but typically test thei...
Detecting out-of-distribution (OOD) inputs is a central challenge for safely deploying machine learn...
Out-of-distribution (OOD) detection is important for machine learning models deployed in the wild. R...
Large pre-trained models such as CLIP or ALIGN offer consistent accuracy across a range of data dist...
Robustness to natural distribution shifts has seen remarkable progress thanks to recent pre-training...
This paper presents a novel evaluation framework for Out-of-Distribution (OOD) detection that aims t...
Modern machine learning techniques have demonstrated their excellent capabilities in many areas. Des...
In machine learning, we traditionally evaluate the performance of a single model, averaged over a co...
When deployed in the real world, machine learning models inevitably encounter changes in the data di...
Out-of-distribution detection is a common issue in deploying vision models in practice and solving i...
Machine learning algorithms typically assume that training and test examples are drawn from the same...
Unsupervised out-of-distribution (U-OOD) detection has recently attracted much attention due its imp...
While deep learning models have seen widespread success in controlled environments, there are still ...
The ability to detect Out-of-Distribution (OOD) data is important in safety-critical applications of...
In the problem of out-of-distribution (OOD) detection, the usage of auxiliary data as outlier data f...
Existing works have made great progress in improving adversarial robustness, but typically test thei...
Detecting out-of-distribution (OOD) inputs is a central challenge for safely deploying machine learn...
Out-of-distribution (OOD) detection is important for machine learning models deployed in the wild. R...
Large pre-trained models such as CLIP or ALIGN offer consistent accuracy across a range of data dist...
Robustness to natural distribution shifts has seen remarkable progress thanks to recent pre-training...
This paper presents a novel evaluation framework for Out-of-Distribution (OOD) detection that aims t...
Modern machine learning techniques have demonstrated their excellent capabilities in many areas. Des...
In machine learning, we traditionally evaluate the performance of a single model, averaged over a co...
When deployed in the real world, machine learning models inevitably encounter changes in the data di...
Out-of-distribution detection is a common issue in deploying vision models in practice and solving i...
Machine learning algorithms typically assume that training and test examples are drawn from the same...
Unsupervised out-of-distribution (U-OOD) detection has recently attracted much attention due its imp...
While deep learning models have seen widespread success in controlled environments, there are still ...