SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2212.09000 · 2022

Confidence-aware Training of Smoothed Classifiers for Certified Robustness

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 6 functions out of this paper's own repositories and ran 5 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
alinlab/smoothing-catrs canonical 5 of 6
FunctionStatusWhere it lives
consistency_loss Ran alinlab/smoothing-catrs/code/consistency.py
code served (permissive licence) · get_code("3d45147f0a674bc0")
conv3x3 Ran alinlab/smoothing-catrs/code/archs/cifar_resnet.py
code served (permissive licence) · get_code("fac5364e2f53c6db")
entropy Ran alinlab/smoothing-catrs/code/consistency.py
code served (permissive licence) · get_code("23f822c3d6d64a12")
get_num_classes Ran alinlab/smoothing-catrs/code/datasets.py
code served (permissive licence) · get_code("ddca6b8fb149d7f9")
kl_div Ran alinlab/smoothing-catrs/code/consistency.py
code served (permissive licence) · get_code("6518bcfd499a3632")
get_dataset Not yet run alinlab/smoothing-catrs/code/datasets.py
code served (permissive licence) · get_code("8a021efe12dc7d06")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Any classifier can be "smoothed out" under Gaussian noise to build a new classifier that is provably robust to $\ell_2$-adversarial perturbations, viz., by averaging its predictions over the noise via randomized smoothing. Under the smoothed classifiers, the fundamental trade-off between accuracy and (adversarial) robustness has been well evidenced in the literature: i.e., increasing the robustness of a classifier for an input can be at the expense of decreased accuracy for some other inputs. In this paper, we propose a simple training method leveraging this trade-off to obtain robust smoothed classifiers, in particular, through a sample-wise control of robustness over the training samples. We make this control feasible by using "accuracy under Gaussian noise" as an easy-to-compute proxy of adversarial robustness for an input. Specifically, we differentiate the training objective depending on this proxy to filter out samples that are unlikely to benefit from the worst-case (adversarial) objective. Our experiments show that the proposed method, despite its simplicity, consistently exhibits improved certified robustness upon state-of-the-art training methods. Somewhat surprisingly, we find these improvements persist even for other notions of robustness, e.g., to various types of common corruptions.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2212.09000")
get_code_for_paper("2212.09000")
have("2212.09000")

Connect an agent — have() is free.