SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1808.08444 · 2018

Guiding Deep Learning System Testing using Surprise Adequacy

arXiv · PDF · Open in the Atlas

Code that ran

4 of 4 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 1 has no contract verdict; 3 failures of our instrument; 0 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/1808.08444.json · badge: /api/ran/1808.08444.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
coinse/sadl canonical (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
thouyang/ccd-dsa extension (classifier label)
classifier basis: classifier_2025
2 ran · 0 no recorded run
testingautomated-usi/surprise-adequacy community (classifier label)
classifier basis not recorded on the link
1 ran · 0 no recorded run
FunctionStatusWhere it lives
cal_cla_matrix Ran
ran; our contract draft was wrong, not the code
thouyang/ccd-dsa/dif_dsa.py
pointer only (licence: NONE) · get_code("06f12f4f35bb3d31")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, thouyang/ccd-dsa when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

find_closest_at Ran
ran; our contract draft was wrong, not the code
thouyang/ccd-dsa/dif_dsa.py
pointer only (licence: NONE) · get_code("a95b762135227755")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, thouyang/ccd-dsa when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

find_closest_at Ran
ran; our contract draft was wrong, not the code
coinse/sadl/sa.py
code served (permissive licence) · get_code("24a57530f2dce7e0")
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) and returned tuple. The image it ran in was not recorded. If your copy returns the same type 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 sample activation traces
# at: a single activation trace (1D array)
# train_ats: multiple activation traces (2D array where each row is a trace)

np.random.seed(0)

# Single activation trace from input
at = np.random.randn(4)

# Training activation traces - 3 samples, each with 4 dimensions
train_ats = np.random.randn(3, 4)

ARGS = {
    "at": at,
    "train_ats": train_ats,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
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
train_model Ran
ran on a synthesized input
testingautomated-usi/surprise-adequacy/case_study/mnist.py
code served (permissive licence) · get_code("da4d212624637316")
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 tuple, ndim 1, 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
import tensorflow as tf

# Mock the MNIST dataset to avoid downloading
def mock_mnist_load():
    x_train = np.random.randint(0, 256, (100, 28, 28)).astype('uint8')
    y_train = np.random.randint(0, 10, (100,))
    x_test = np.random.randint(0, 256, (20, 28, 28)).astype('uint8')
    y_test = np.random.randint(0, 10, (20,))
    return (x_train, y_train), (x_test, y_test)

# Patch the MNIST loader
original_load = tf.keras.datasets.mnist.load_data
tf.keras.datasets.mnist.load_data = mock_mnist_load

ARGS = {
    "model_id": "test_model_v1"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family float — a tuple output has no array shape of its own; none is stated
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

Also present in other papers' code

Of 4 samples lifted from this paper's linked repositories, 4 carry a function identity and 0 do not. Of the 4 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 4 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

Deep Learning (DL) systems are rapidly being adopted in safety and security critical domains, urgently calling for ways to test their correctness and robustness. Testing of DL systems has traditionally relied on manual collection and labelling of data. Recently, a number of coverage criteria based on neuron activation values have been proposed. These criteria essentially count the number of neurons whose activation during the execution of a DL system satisfied certain properties, such as being above predefined thresholds. However, existing coverage criteria are not sufficiently fine grained to capture subtle behaviours exhibited by DL systems. Moreover, evaluations have focused on showing correlation between adversarial examples and proposed criteria rather than evaluating and guiding their use for actual testing of DL systems. We propose a novel test adequacy criterion for testing of DL systems, called Surprise Adequacy for Deep Learning Systems (SADL), which is based on the behaviour of DL systems with respect to their training data. We measure the surprise of an input as the difference in DL system's behaviour between the input and the training data (i.e., what was learnt during training), and subsequently develop this as an adequacy criterion: a good test input should be sufficiently but not overtly surprising compared to training data. Empirical evaluation using a range of DL systems from simple image classifiers to autonomous driving car platforms shows that systematic sampling of inputs based on their surprise can improve classification accuracy of DL systems against adversarial examples by up to 77.5% via retraining.

For agents

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

get_harvested_code_for_paper("1808.08444")
get_code_for_paper("1808.08444")
have("1808.08444")

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

curl https://syntology.ai/api/ran/1808.08444.json

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

[![Syntology run record](https://syntology.ai/api/ran/1808.08444.svg)](https://syntology.ai/paper/1808.08444)

Connect an agent — have() is free.