SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1809.03113 · 2018

Certified Adversarial Robustness with Additive Noise

arXiv · PDF · Open in the Atlas

Code that ran

2 of 9 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 2 have no contract verdict; 0 failures of our instrument; 7 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/1809.03113.json · badge: /api/ran/1809.03113.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
Bai-Li/STN-Code canonical (classifier label)
classifier basis: census
2 ran · 7 no recorded run
FunctionStatusWhere it lives
error_rate Ran
ran on a synthesized input
Bai-Li/STN-Code/tf_utils_adv.py
code served (permissive licence) · get_code("1ab9b1ec5119608a")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned float64, ndim 0, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Create small predictions and labels arrays
# predictions: shape (2, 4) - batch size 2, 4 classes
# labels: shape (2, 4) - batch size 2, 4 classes (one-hot encoded)
np.random.seed(0)

predictions = np.array([
    [0.1, 0.2, 0.6, 0.1],
    [0.3, 0.5, 0.1, 0.1]
], dtype=np.float32)

labels = np.array([
    [0, 0, 1, 0],
    [0, 1, 0, 0]
], dtype=np.float32)

ARGS = {
    "predictions": predictions,
    "labels": labels
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
linf_loss Ran
ran on a synthesized input
Bai-Li/STN-Code/attack_utils.py
code served (permissive licence) · get_code("376cff0722b80724")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 1, dtype family float, shape (2,) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# The function linf_loss takes two arguments X1 and X2
# It computes the maximum absolute difference along axes (1, 2, 3)
# This implies 4D arrays/tensors with shape (batch, dim1, dim2, dim3)

# Create two small 4D numpy arrays with distinct dimensions
batch_size = 2
dim1 = 4
dim2 = 3
dim3 = 5

X1 = np.random.randn(batch_size, dim1, dim2, dim3).astype(np.float32)
X2 = np.random.randn(batch_size, dim1, dim2, dim3).astype(np.float32)

ARGS = {
    "X1": X1,
    "X2": X2,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family float, shape (2,) — shape (2,) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
batch_eval Not yet run
unverified (harvested, no recorded run)
Bai-Li/STN-Code/tf_utils_adv.py
code served (permissive licence) · get_code("a7ed2e3707617703")
How it ran

no invocation recorded.

gen_adv_loss Not yet run
unverified (harvested, no recorded run)
Bai-Li/STN-Code/attack_utils.py
code served (permissive licence) · get_code("0d675bf30467b0db")
How it ran

no invocation recorded.

gen_grad Not yet run
unverified (harvested, no recorded run)
Bai-Li/STN-Code/attack_utils.py
code served (permissive licence) · get_code("3da24a5adc2ae2b4")
How it ran

no invocation recorded.

load_model_mnist Not yet run
unverified (harvested, no recorded run)
Bai-Li/STN-Code/mnist.py
code served (permissive licence) · get_code("ce08fb5409bf3bb8")
How it ran

no invocation recorded.

symbolic_alpha_fgs Not yet run
unverified (harvested, no recorded run)
Bai-Li/STN-Code/fgs.py
code served (permissive licence) · get_code("cabead92fdbdf14b")
How it ran

no invocation recorded.

symbolic_fgs Not yet run
unverified (harvested, no recorded run)
Bai-Li/STN-Code/fgs.py
code served (permissive licence) · get_code("be178859da767042")
How it ran

no invocation recorded.

tf_test_acc Not yet run
unverified (harvested, no recorded run)
Bai-Li/STN-Code/tf_utils_adv.py
code served (permissive licence) · get_code("51d6534c3e5260b5")
How it ran

no invocation recorded.

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 9 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

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

Abstract

The existence of adversarial data examples has drawn significant attention in the deep-learning community; such data are seemingly minimally perturbed relative to the original data, but lead to very different outputs from a deep-learning algorithm. Although a significant body of work on developing defensive models has been considered, most such models are heuristic and are often vulnerable to adaptive attacks. Defensive methods that provide theoretical robustness guarantees have been studied intensively, yet most fail to obtain non-trivial robustness when a large-scale model and data are present. To address these limitations, we introduce a framework that is scalable and provides certified bounds on the norm of the input manipulation for constructing adversarial examples. We establish a connection between robustness against adversarial perturbation and additive random noise, and propose a training strategy that can significantly improve the certified bounds. Our evaluation on MNIST, CIFAR-10 and ImageNet suggests that the proposed method is scalable to complicated models and large data sets, while providing competitive robustness to state-of-the-art provable defense methods.

For agents

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

get_harvested_code_for_paper("1809.03113")
get_code_for_paper("1809.03113")
have("1809.03113")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/1809.03113.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/1809.03113.svg)](https://syntology.ai/paper/1809.03113)

Connect an agent — have() is free.