SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1810.12715 · 2018

On the Effectiveness of Interval Bound Propagation for Training Verifiably Robust Models

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 6 functions out of this paper's own repositories and ran 1 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
copy not recorded — 1 of 3
sungyoon-lee/bcp pwc_unofficial 0 of 3
FunctionStatusWhere it lives
catNonNullErrors Ran this paper's copy was not recorded; identical code first harvested from eth-sri/diffai
pointer only · get_code("f8f37cd8a52870c1")
argparser Not yet run sungyoon-lee/bcp/utils.py
code served (permissive licence) · get_code("846b8ce1a0cef2b3")
creluBoxy Not yet run this paper's copy was not recorded; identical code first harvested from eth-sri/diffai
pointer only · get_code("a49c0dd16152c346")
creluBoxySound Not yet run this paper's copy was not recorded; identical code first harvested from eth-sri/diffai
pointer only · get_code("c42ef7b9012126aa")
model_cnn_4layer Not yet run sungyoon-lee/bcp/utils.py
code served (permissive licence) · get_code("fb45ae0df896c8db")
select_model Not yet run sungyoon-lee/bcp/utils.py
code served (permissive licence) · get_code("bb465636abefa4b4")

Repositories linked to this paper

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

Abstract

Recent work has shown that it is possible to train deep neural networks that are provably robust to norm-bounded adversarial perturbations. Most of these methods are based on minimizing an upper bound on the worst-case loss over all possible adversarial perturbations. While these techniques show promise, they often result in difficult optimization procedures that remain hard to scale to larger networks. Through a comprehensive analysis, we show how a simple bounding technique, interval bound propagation (IBP), can be exploited to train large provably robust neural networks that beat the state-of-the-art in verified accuracy. While the upper bound computed by IBP can be quite weak for general networks, we demonstrate that an appropriate loss and clever hyper-parameter schedule allow the network to adapt such that the IBP bound is tight. This results in a fast and stable learning algorithm that outperforms more sophisticated methods and achieves state-of-the-art results on MNIST, CIFAR-10 and SVHN. It also allows us to train the largest model to be verified beyond vacuous bounds on a downscaled version of ImageNet.

For agents

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

get_harvested_code_for_paper("1810.12715")
get_code_for_paper("1810.12715")
have("1810.12715")

Connect an agent — have() is free.