Papers › Efficient Neural Network Robustness Certification with General Activation Functions

Efficient Neural Network Robustness Certification with General Activation Functions

2 Nov 2018NeurIPS 2018 12arXiv:1811.00866archive 2025-07-28

Huan Zhang, Tsui-Wei Weng, Pin-Yu Chen, Cho-Jui Hsieh, Luca Daniel

Finding minimum distortion of adversarial examples and thus certifying robustness in neural network classifiers for given data points is known to be a challenging problem. Nevertheless, recently it has been shown to be possible to give a non-trivial certified lower bound of minimum adversarial distortion, and some recent progress has been made towards this direction by exploiting the piece-wise linear nature of ReLU activations. However, a generic robustness certification for general activation functions still remains largely unexplored. To address this issue, in this paper we introduce CROWN, a general framework to certify robustness of neural networks with general activation functions for given input data points. The novelty in our algorithm consists of bounding a given activation function with linear and quadratic functions, hence allowing it to tackle general activation functions including but not limited to four popular choices: ReLU, tanh, sigmoid and arctan. In addition, we facilitate the search for a tighter certified lower bound by adaptively selecting appropriate surrogates for each neuron activation. Experimental results show that CROWN on ReLU networks can notably improve the certified lower bounds compared to the current state-of-the-art algorithm Fast-Lin, while having comparable computational efficiency. Furthermore, CROWN also demonstrates its effectiveness and flexibility on networks with general activation functions, including tanh, sigmoid and arctan.

PaperPDFConference PDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

No sample from the archive's official repository (huanzhang12/CROWN-Robustness-Certification) is linked to this paper; the 43 samples below come from 5 community repositories. Syntology harvested code from huanzhang12/CROWN-Robustness-Certification for another paper.

huanzhang12/CROWN-Robustness-Certification: Syntology's graph links these samples to Provable defenses against adversarial examples via the convex outer adversarial polytope (arXiv:1711.00851; that paper's own run record: 21 ran (of which 15 constructed an object rather than computing a result; 19 with no instrument failure: 1 honoured, 1 violated, 17 with no contract checked; 2 where Syntology's instrument failed) · 23 unverified); nothing checks that they implement this paper's method.

Syntology 43 code samples harvested from 5 repositories linked to this paper: 33 ran (of which 0 constructed an object rather than computing a result; 31 with no instrument failure: 2 honoured, 0 violated, 29 with no contract checked; 2 where Syntology's instrument failed) · 10 unverified. Where Syntology's instrument failed: 2 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 43 samples from 5 repositories, 33 ran (31 with no instrument failure; 2 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

No sample is linked to this paper from 7 of its repositories that Syntology harvested for other papers:

Levels written between 2026-09-22 and 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

16 repositories listed; official and paper-mentioned ones first.

huanzhang12/CROWN-Robustness-Certification officialmentioned in papermentioned on GitHubtfSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Provable defenses against adversarial examples via the convex outer adversarial polytope (arXiv:1711.00851; that paper's own run record: 21 ran (of which 15 constructed an object rather than computing a result; 19 with no instrument failure: 1 honoured, 1 violated, 17 with no contract checked; 2 where Syntology's instrument failed) · 23 unverified); nothing checks that they implement this paper's method. report
AkhilanB/CNN-Cert mentioned on GitHubtfSyntology: 2 samples, 1 ran (1 instrument-failed) · 1 unverified report
CROWN-Robustness/Crown mentioned on GitHubtfNOASSERTIONSyntology: not harvested report
IBM/CROWN-Robustness-Certification mentioned on GitHubtfApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Tight Certification of Adversarially Trained Neural Networks via Nonconvex Low-Rank... (arXiv:2211.17244; that paper's own run record: 18 ran (of which 0 constructed an object rather than computing a result; 18 with no instrument failure: 2 honoured, 0 violated, 16 with no contract checked; 0 where Syntology's instrument failed) · 6 unverified); nothing checks that they implement this paper's method. report
Verified-Intelligence/alpha-beta-CROWN_vnncomp2024 mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Fast and Complete: Enabling Complete Neural Network Verification with Rapid and... (arXiv:2011.13824; that paper's own run record: 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified), Beta-CROWN: Efficient Bound Propagation with Per-neuron Split Constraints for Complete... (arXiv:2103.06624; that paper's own run record: 4 ran (of which 1 constructed an object rather than computing a result; 3 with no instrument failure: 1 honoured, 1 violated, 1 with no contract checked; 1 where Syntology's instrument failed) · 1 unverified), General Cutting Planes for Bound-Propagation-Based Neural Network Verification (arXiv:2208.05740; that paper's own run record: 2 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 1 where Syntology's instrument failed) · 4 unverified) and 1 more (every one in this page's JSON twin); nothing checks that they implement this paper's method. report
ZhaoyangLyu/FROWN mentioned on GitHubpytorchSyntology: 2 samples, 1 ran (1 instrument-failed) · 1 unverified report
aria-systems-group/formal-deep-kernel-synthesis mentioned on GitHubpytorchSyntology: 1 sample, 0 ran · 1 unverified report
avrahamraviv/formal-verification-of-object-detection mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to General Cutting Planes for Bound-Propagation-Based Neural Network Verification (arXiv:2208.05740; that paper's own run record: 2 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 1 where Syntology's instrument failed) · 4 unverified); nothing checks that they implement this paper's method. report
huanzhang12/CROWN-IBP mentioned on GitHubpytorchBSD-2-ClauseSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Automatic Perturbation Analysis for Scalable Certified Robustness and Beyond (arXiv:2002.12920; that paper's own run record: 14 ran (of which 0 constructed an object rather than computing a result; 12 with no instrument failure: 0 honoured, 3 violated, 9 with no contract checked; 2 where Syntology's instrument failed) · 4 unverified); nothing checks that they implement this paper's method. report
huanzhang12/CertifiedReLURobustness mentioned on GitHubtfBSD-2-ClauseSyntology: 13 samples, 11 ran (0 instrument-failed) · 2 unverified report
huanzhang12/RecurJac mentioned on GitHubtfBSD-2-ClauseSyntology: not harvested report
huanzhang12/RecurJac-Jacobian-Bounds mentioned on GitHubtfBSD-2-ClauseSyntology: 25 samples, 20 ran (0 instrument-failed) · 5 unverified report
huanzhang12/RecurJac-and-CROWN mentioned on GitHubtfBSD-2-ClauseSyntology: not harvested report
huanzhang12/alpha-beta-CROWN mentioned on GitHubpytorchNOASSERTIONSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Neural Network Verification with Branch-and-Bound for General Nonlinearities (arXiv:2405.21063; that paper's own run record: 11 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 0 violated, 3 with no contract checked; 7 where Syntology's instrument failed) · 3 unverified); nothing checks that they implement this paper's method. report
huanzhang12/sa_ppo mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Robust Reinforcement Learning on State Observations with Learned Optimal Adversary (arXiv:2101.08452; that paper's own run record: 3 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 3 where Syntology's instrument failed) · 5 unverified); nothing checks that they implement this paper's method. report
verified-intelligence/alpha-beta-crown mentioned on GitHubpytorchNOASSERTIONSyntology: not harvested report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

43 samples harvested: 33 ran (of which 0 constructed an object rather than computing a result; 31 with no instrument failure: 2 honoured, 0 violated, 29 with no contract checked; 2 where Syntology's instrument failed) · 10 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

2ran · honoured contract
2ran · our draft was wrong
29ran
10unverified

Licence: 3 of the 43 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 5 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity.

AkhilanB/CNN-Cert community (archive-listed) 2 samples, 1 ran (1 instrument-failed) · 1 unverified
command pymain.py ran · our draft was wrong fingerprinted BSD-2-Clause (permissive) · 961f8ae9024624c4 · report fixture: signature cmd:str/0/str · returned str · level recorded 2026-09-22
run in 143 papers pymain.py unverified BSD-2-Clause (permissive) · fd87678657cdbe5f · report
ZhaoyangLyu/FROWN community (archive-listed) 2 samples, 1 ran (1 instrument-failed) · 1 unverified
main in 235 papers train_model.py ran · our draft was wrong no licence file found · pointer only · 05f24c1d58d4690a · report fixture: signature model:callable/0/?, train_loader:DataLoader/0/?, test_loader:DataLoader/0/?, device:device/0/?, epochs:int/0/int, parallel:bool/0/bool, lr:float/0/float, lr_decay_interval:int/0/int, lr_decay_factor:float/0/float, log_interval:int/0/int · returned SimpleModel · level recorded 2026-09-22
getMaximumEps terminal_runner_certify_targeted_attack.py unverified no licence file found · pointer only · 78bdba96d37bb59e · report
aria-systems-group/formal-deep-kernel-synthesis community (archive-listed) 1 sample, 0 ran · 1 unverified
beta_bias alpha-beta-CROWN/auto_LiRPA/beta_crown.py unverified no licence file found · pointer only · 9270bd3e922adfe2 · report
huanzhang12/CertifiedReLURobustness community (archive-listed) 13 samples, 11 ran (0 instrument-failed) · 2 unverified
ReLU in 8 papers get_bounds_ours.py ran fingerprinted BSD-2-Clause (permissive) · 753617a588e4fc41 · report fixture: signature vec:ndarray/1/float · returned ndarray · level recorded 2026-09-25
extract_data in 24 papers setup_mnist.py ran · honoured contract BSD-2-Clause (permissive) · 86c9fa246614e9a9 · report fixture: signature filename:str/0/str, num_images:int/0/int · returned ndarray · level recorded 2026-09-22
extract_labels in 25 papers setup_mnist.py ran · honoured contract BSD-2-Clause (permissive) · a0cf6bb81ff7819e · report fixture: returned ndarray · level recorded 2026-09-22
format_norm genlatex.py ran fingerprinted BSD-2-Clause (permissive) · 32794458cec639b5 · report fixture: signature n:float/0/float · returned str · level recorded 2026-09-25
format_time in 140 papers genlatex.py ran fingerprinted BSD-2-Clause (permissive) · 0e83f0ab852f5a0b · report fixture: signature t:float/0/float · returned str · level recorded 2026-09-25
get_layer_bound get_bounds_ours.py ran BSD-2-Clause (permissive) · 37fee45ec17c90a3 · report fixture: signature W_Nk:ndarray/2/float, b_Nk:ndarray/1/float, UB_prev:ndarray/1/float, LB_prev:ndarray/1/float, is_last:bool/0/bool, x0:ndarray/1/float, eps:float/0/float, p_n:int/0/int · returned tuple · level recorded 2026-09-25
get_weights_list get_bounds_ours.py ran BSD-2-Clause (permissive) · fcf70a46a28d44a2 · report fixture: signature model:callable/0/? · returned tuple · level recorded 2026-09-25
l1_dist utils.py ran fingerprinted BSD-2-Clause (permissive) · 754f9759bf35afba · report fixture: signature x:ndarray/2/float, y:ndarray/2/float · returned float64 · level recorded 2026-09-25
l2_dist in 7 papers utils.py ran fingerprinted BSD-2-Clause (permissive) · c616cfbcc11326f3 · report fixture: signature x:ndarray/2/float, y:ndarray/2/float · returned float64 · level recorded 2026-09-25
linf_dist utils.py ran fingerprinted BSD-2-Clause (permissive) · 4fc767413887cc02 · report fixture: signature x:ndarray/2/float, y:ndarray/2/float · returned float64 · level recorded 2026-09-25
spectral_bound get_bounds_others.py ran BSD-2-Clause (permissive) · 82879d94252d9441 · report fixture: signature weights:list/1/ndarray, biases:list/1/ndarray, pred_label:int/0/int, target_label:int/0/int, x0:ndarray/1/float, predictions:ndarray/1/float, numlayer:int/0/int, p:str/0/str, untargeted:bool/0/bool · returned float64 · level recorded 2026-09-25
load_batch in 18 papers setup_cifar.py unverified BSD-2-Clause (permissive) · ab7faf0047f2c872 · report
load_batch in 18 papers setup_cifar.py unverified BSD-2-Clause (permissive) · 3b4fb87c66244ebf · report
huanzhang12/RecurJac-Jacobian-Bounds community (archive-listed) 25 samples, 20 ran (0 instrument-failed) · 5 unverified
ReLU in 8 papers bound_base.py ran fingerprinted BSD-2-Clause (permissive) · 67421da32aae154e · report fixture: signature vec:ndarray/1/float · returned ndarray · level recorded 2026-09-25
add_cache cachefy.py ran fingerprinted BSD-2-Clause (permissive) · 35286dc13357efe4 · report fixture: signature l:str/0/str · returned str · level recorded 2026-09-25
fastlip_2layer bound_fastlin_fastlip.py ran BSD-2-Clause (permissive) · bbc8cb9f0698c825 · report fixture: signature W2:ndarray/2/float, W1:ndarray/2/float, neuron_state:ndarray/1/int, norm:int/0/int · returned tuple · level recorded 2026-09-25
get_area in 10 papers activation_functions_quad.py ran BSD-2-Clause (permissive) · 90f4d0f321d7321a · report fixture: signature a:float/0/float, b:float/0/float, c:float/0/float, lower:float/0/float, upper:float/0/float · returned float · level recorded 2026-09-25
get_lower_quad_parameterized_lgtu activation_functions_quad.py ran fingerprinted BSD-2-Clause (permissive) · 859c42b9a1c7fb44 · report fixture: signature u:float/0/float, l:float/0/float, theta:float/0/float · returned tuple · level recorded 2026-09-25
get_model_meta_real mnist_cifar_models.py ran BSD-2-Clause (permissive) · 63f0738b7f84b19d · report fixture: signature filename:str/0/str, model_loader:callable/0/? · returned tuple · level recorded 2026-09-25
get_upper_quad_parameterized activation_functions_quad.py ran fingerprinted BSD-2-Clause (permissive) · 171e3d9dae850e45 · report fixture: signature u:float/0/float, l:float/0/float, theta:float/0/float · returned tuple · level recorded 2026-09-25
get_weights_list bound_base.py ran BSD-2-Clause (permissive) · 19a9b690599869b4 · report fixture: signature model:callable/0/? · returned tuple · level recorded 2026-09-25
init_fastlin_bounds bound_fastlin_fastlip.py ran BSD-2-Clause (permissive) · 98c1b7034a8af60a · report fixture: signature Ws:list/1/ndarray · returned list · level recorded 2026-09-25
interval_bound bound_interval.py ran BSD-2-Clause (permissive) · 1e47ec700db6f2be · report fixture: signature W_Nk:ndarray/2/float, b_Nk:ndarray/1/float, UB_prev:ndarray/1/float, LB_prev:ndarray/1/float, x0:ndarray/1/float, eps:float/0/float, p_n:float/0/float · returned tuple · level recorded 2026-09-25
leaky_relu_ub_pn activation_functions.py ran fingerprinted BSD-2-Clause (permissive) · bcfb10a6fdde6ee1 · report fixture: signature u:float/0/float, l:float/0/float, k:float/0/float · returned tuple · level recorded 2026-09-25
proj_l1 bound_crown_quad.py ran fingerprinted BSD-2-Clause (permissive) · 2dfae576a64727fc · report fixture: signature x_real:ndarray/1/float, x0:ndarray/1/float, q:float/0/float · returned ndarray · level recorded 2026-09-25
proj_l2 bound_crown_quad.py ran fingerprinted BSD-2-Clause (permissive) · dd8e7c5361d6a51b · report fixture: signature x:ndarray/1/float, x0:ndarray/1/float, eps:float/0/float · returned ndarray · level recorded 2026-09-25
proj_li bound_crown_quad.py ran BSD-2-Clause (permissive) · cba73c35959ddc1d · report fixture: signature x:ndarray/1/float, i_lb:ndarray/0/float, i_ub:ndarray/0/float · returned ndarray · level recorded 2026-09-25
recurjac_bound_2layer bound_recurjac.py ran BSD-2-Clause (permissive) · de5a9dcd60ab18e8 · report fixture: signature W2:ndarray/2/float, W1:ndarray/2/float, grad_UB:ndarray/1/float, grad_LB:ndarray/1/float · returned tuple · level recorded 2026-09-25
relu_grad_bounds bound_recurjac.py ran fingerprinted BSD-2-Clause (permissive) · fa98765e8a5ffef2 · report fixture: signature lb:float/0/float, ub:float/0/float, alpha:float/0/float · returned tuple · level recorded 2026-09-25
relu_ub_pn activation_functions.py ran fingerprinted BSD-2-Clause (permissive) · 88ac5c47845ee90e · report fixture: signature u:ndarray/1/float, l:ndarray/1/float · returned tuple · level recorded 2026-09-25
remove_cache cachefy.py ran fingerprinted BSD-2-Clause (permissive) · 0b2c8166cb0eec6b · report fixture: signature l:str/0/str · returned str · level recorded 2026-09-25
sigmoid_grad_bounds bound_recurjac.py ran BSD-2-Clause (permissive) · fe4cf0885a1400ce · report fixture: signature lb:float/0/float, ub:float/0/float, grad_f:callable/0/? · returned tuple · level recorded 2026-09-25
spectral_bound bound_spectral.py ran BSD-2-Clause (permissive) · 22f9598f566ecc55 · report fixture: signature weights:list/1/ndarray, biases:list/1/ndarray, pred_label:int/0/int, target_label:int/0/int, x0:ndarray/1/float, predictions:ndarray/1/float, numlayer:int/0/int, activation:str/0/str, p:float/0/float, untargeted:bool/0/bool · returned tuple · level recorded 2026-09-25
get_general_bounds bound_crown.py unverified BSD-2-Clause (permissive) · 91b496effee73d52 · report
get_leaky_relu_bounds bound_crown.py unverified BSD-2-Clause (permissive) · 9047985cc861bb05 · report
get_model_meta mnist_cifar_models.py unverified BSD-2-Clause (permissive) · aff1aec26ddd75e4 · report
get_relu_bounds bound_crown.py unverified BSD-2-Clause (permissive) · d338c893ca81dac0 · report
relu_lb_pn activation_functions.py unverified BSD-2-Clause (permissive) · 52125b4504b9b441 · report

Tasks

Computational EfficiencyEfficient Neural Network

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

ReLU

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections