SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2102.05188 · ICLR · 2021

CaPC Learning: Confidential and Private Collaborative Learning

Nicolas Papernot, Christopher Choquette-Choo, Natalie Dullerud, Adam Dziedzic, Yunxiang Zhang, Somesh Jha, Xiao Wang

arXiv · PDF · Open in the Atlas

Code that ran

6 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, 1 violated it and 0 have no contract verdict; 5 failures of our instrument; 3 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/2102.05188.json · badge: /api/ran/2102.05188.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
cleverhans-lab/capc-iclr canonical (classifier label)
classifier basis: classifier_2025
6 ran · 3 no recorded run
FunctionStatusWhere it lives
array_str Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
cleverhans-lab/capc-iclr/he_server.py
code served (permissive licence) · get_code("ca29535886dce2f6")
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 str. 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 torch
import numpy as np

torch.manual_seed(0)

# The function array_str takes two parameters:
# - x: an array-like object (should support .shape attribute and iteration)
# - out: a string for accumulation (default='')
# 
# The function recursively processes multi-dimensional arrays and converts them to string representation.

# Create a small 2D tensor to test the function
x = torch.tensor([[1, 2, 3, 4],
                   [5, 6, 7, 8]], dtype=torch.float32)

ARGS = {
    'x': x,
    'out': ''
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not 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
get_models Ran
ran; our contract draft was wrong, not the code
cleverhans-lab/capc-iclr/run_experiment.py
code served (permissive licence) · get_code("1aac2bdd0fc20a12")
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 os
import tempfile

# Create a temporary directory with some model files
temp_dir = tempfile.mkdtemp()

# Create dummy model files
for i in range(3):
    with open(os.path.join(temp_dir, f"model_{i}.pt"), "w") as f:
        f.write("dummy model content")

ARGS = {
    "model_dir": temp_dir,
    "n_parties": 3,
    "ignore_parties": False,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
none outside the standard library — 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
get_rstar_server Ran
ran; our contract draft was wrong, not the code
cleverhans-lab/capc-iclr/he_server.py
code served (permissive licence) · get_code("4ca6cbd29699e632")
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 2, dtype family float, shape (2, 4) 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

max_logit = np.random.randn(2, 4).astype(np.float32)
batch_size = 2
num_classes = 4
exp = 2

ARGS = {
    "max_logit": max_logit,
    "batch_size": batch_size,
    "num_classes": num_classes,
    "exp": exp
}
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 2, dtype family float, shape (2, 4) — shape (2, 4) 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
load_pb_file Ran
ran; our contract draft was wrong, not the code
cleverhans-lab/capc-iclr/he_server.py
code served (permissive licence) · get_code("b577d0ec0af61398")
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 GraphDef. 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 tensorflow as tf
import tempfile
import os

# Create a temporary protobuf file for testing
def create_test_pb_file():
    # Create a simple graph
    graph = tf.Graph()
    with graph.as_default():
        x = tf.constant(1.0, name='input')
        y = tf.multiply(x, 2.0, name='output')
    
    # Convert to graph_def
    graph_def = graph.as_graph_def()
    
    # Write to temporary file
    temp_file = tempfile.NamedTemporaryFile(delete=False, suffix='.pb')
    temp_file.write(graph_def.SerializeToString())
    temp_file.close()
    
    return temp_file.name

# Create the test file
pb_filename = create_test_pb_file()

# Define arguments
ARGS = {
    'filename': pb_filename
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type GraphDef — a GraphDef output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a GraphDef output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not 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
round_array Ran
ran, violated the contract we drafted
cleverhans-lab/capc-iclr/he_server.py
code served (permissive licence) · get_code("9f8912a4ea9ac1d4")
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 2, dtype family int, shape (2, 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

# Construct test inputs for round_array
x_input = np.array([[0.28, 0.32], [0.15, 0.95]])
exp_input = 10

ARGS = {
    "x": x_input,
    "exp": exp_input,
}
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 2, dtype family int, shape (2, 2) — shape (2, 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
server_config_from_flags Ran
ran; our contract draft was wrong, not the code
cleverhans-lab/capc-iclr/he_server.py
code served (permissive licence) · get_code("0766946de8984a9b")
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 ConfigProto. 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 tensorflow as tf
from tensorflow.core.protobuf import rewriter_config_pb2

class FlagsStandIn:
    def __init__(self):
        self.backend = "CPU"
        self.encryption_parameters = "test_params"
        self.enable_client = False
        self.enable_gc = True
        self.mask_gc_inputs = False
        self.mask_gc_outputs = False
        self.num_gc_threads = 4
        self.port = 50051
        self.encrypt_server_data = False
        self.pack_data = False

FLAGS = FlagsStandIn()
tensor_param_name = "test_tensor_param"

ARGS = {
    "FLAGS": FLAGS,
    "tensor_param_name": tensor_param_name,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ConfigProto — a ConfigProto output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ConfigProto output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not 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
load_mnist_data Not yet run
unverified (harvested, no recorded run)
cleverhans-lab/capc-iclr/he_server.py
code served (permissive licence) · get_code("cdfbf8e12fda903f")
How it ran

no invocation recorded.

print_nodes Not yet run
unverified (harvested, no recorded run)
cleverhans-lab/capc-iclr/he_server.py
code served (permissive licence) · get_code("7c0c257cb3c50d60")
How it ran

no invocation recorded.

run_server Not yet run
unverified (harvested, no recorded run)
cleverhans-lab/capc-iclr/he_server.py
code served (permissive licence) · get_code("5656950c8d2c2223")
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

Machine learning benefits from large training datasets, which may not always be possible to collect by any single entity, especially when using privacy-sensitive data. In many contexts, such as healthcare and finance, separate parties may wish to collaborate and learn from each other's data but are prevented from doing so due to privacy regulations. Some regulations prevent explicit sharing of data between parties by joining datasets in a central location (confidentiality). Others also limit implicit sharing of data, e.g., through model predictions (privacy). There is currently no method that enables machine learning in such a setting, where both confidentiality and privacy need to be preserved, to prevent both explicit and implicit sharing of data. Federated learning only provides confidentiality, not privacy, since gradients shared still contain private information. Differentially private learning assumes unreasonably large datasets. Furthermore, both of these learning paradigms produce a central model whose architecture was previously agreed upon by all parties rather than enabling collaborative learning where each party learns and improves their own local model. We introduce Confidential and Private Collaborative (CaPC) learning, the first method provably achieving both confidentiality and privacy in a collaborative setting. We leverage secure multiparty computation (MPC), homomorphic encryption (HE), and other techniques in combination with privately aggregated teacher models. We demonstrate how CaPC allows participants to collaborate without having to explicitly join their training sets or train a central model. Each party is able to improve the accuracy and fairness of their model, even in settings where each party has a model that performs well on their own dataset or when datasets are not IID and model architectures are heterogeneous across parties. 1

For agents

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

get_harvested_code_for_paper("2102.05188")
get_code_for_paper("2102.05188")
have("2102.05188")

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

curl https://syntology.ai/api/ran/2102.05188.json

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

[![Syntology run record](https://syntology.ai/api/ran/2102.05188.svg)](https://syntology.ai/paper/2102.05188)

Connect an agent — have() is free.