Oracle-oriented Robustness: Robust Image Model Evaluation with Pretrained Models as Surrogate Oracle

Peiyan Zhang, Sunghun Kim, Eric Xing, Haohan Wang

Keywords: robustness, distribution shift, reliable machine learning

Abstract: Machine learning has demonstrated remarkable performances over finite datasets, yet whether the scores over the fixed benchmarks can sufficiently indicate the model’s performances in the real world is still in discussion. In reality, an ideal robust model will probably behave similarly to the oracle (*e.g.*, the human users), thus a good evaluation protocol is probably to evaluate the models’ behaviors in comparison to the oracle. In this paper, we introduce a new robustness measurement that directly measures the image classification model’s performance compared with a surrogate oracle. Besides, we design a simple method that can accomplish the evaluation beyond the scope of the benchmarks. Our method extends the image datasets with new samples that are sufficiently perturbed to be distinct from the ones in the original sets, but are still bounded within the same causal structure the original test image represents, constrained by a surrogate oracle model pretrained with a large amount of samples. As a result, our new method will offer us a new way to evaluate the models’ robustness performances, free of limitations of fixed benchmarks or constrained perturbations, although scoped by the power of the oracle. In addition to the evaluation results, we also leverage our generated data to understand the behaviors of the model and our new evaluation strategies.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Submission Guidelines: Yes

Please Choose The Closest Area That Your Submission Falls Into: Social Aspects of Machine Learning (eg, AI safety, fairness, privacy, interpretability, human-AI interaction, ethics)

TL;DR: We offer a dynamic evaluation protocol that evaluates vision models' robustness across generic i.i.d benchmarks, without requirement on the prior knowledge of the underlying causal structure depicted by the images and additional human efforts.