SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2411.06346 · 2024

Activation Map Compression through Tensor Decomposition for Deep Learning

arXiv · PDF · Open in the Atlas

Code that ran

8 of 8 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; 2 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 6 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/2411.06346.json · badge: /api/ran/2411.06346.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
le-trungnguyen/neurips2024-activationcompression canonical (classifier label)
classifier basis: census
8 ran · 0 no recorded run
FunctionStatusWhere it lives
calculate_mean_and_deviation Ran
ran; our contract draft was wrong, not the code
le-trungnguyen/neurips2024-activationcompression/segmentation/read_result.py
code served (permissive licence) · get_code("99237605df72c193")
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

# The function takes 'data' as a parameter
# It uses: np.mean(data), iterates over data with ((x - mean) ** 2 for x in data),
# len(data), and np.sqrt()
# So data should be an iterable of numeric values

# Create a small dataset
data = [1.0, 2.0, 3.0, 4.0, 5.0]

ARGS = {
    "data": data
}
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 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
count_elements Ran
ran, honoured the contract we drafted
le-trungnguyen/neurips2024-activationcompression/segmentation/train_count_mem.py
code served (permissive licence) · get_code("48d184eb8b65c96d")
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 int, ndim 0, dtype family int, shape () 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 torch

torch.manual_seed(0)

ARGS = {
    'elements': [
        {'type': 'cbr'},
        {'type': 'conv'},
        {'type': 'resnet_basic_block'},
        {'type': 'other'},
        {'type': 'resnet_basic_block'},
    ]
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type int, ndim 0, dtype family int, shape () — shape () 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
extract_config Ran
ran; our contract draft was wrong, not the code
le-trungnguyen/neurips2024-activationcompression/classification/read_result.py
code served (permissive licence) · get_code("fb05a2e34b131f44")
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 os
import tempfile
import torch

# Create a temporary directory with a config.yaml file
temp_dir = tempfile.mkdtemp()
config_path = os.path.join(temp_dir, 'config.yaml')

config_content = """seed_everything: 42
num_of_finetune: 10
backbone: resnet50
name: cifar10
with_SVD: false
with_HOSVD: false
with_grad_filter: false
explained_variance_threshold: 0.95
filt_radius: 3
exp_name: test_experiment
max_epochs: 100
"""

with open(config_path, 'w') as f:
    f.write(config_content)

ARGS = {
    'directory_path': temp_dir
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — 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
extract_config Ran
ran; our contract draft was wrong, not the code
le-trungnguyen/neurips2024-activationcompression/segmentation/read_result.py
code served (permissive licence) · get_code("f8bc83c7ab64df05")
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 os
import tempfile

# Create a temporary directory with a config.yaml file
temp_dir = tempfile.mkdtemp()
config_yaml_path = os.path.join(temp_dir, 'config.yaml')

# Write a valid config.yaml file
config_content = """seed_everything: 42
num_of_finetune: 10
backbone: resnet50
name: cifar10
with_SVD_with_var_compression: false
with_HOSVD_with_var_compression: false
with_grad_filter: false
SVD_var: 0.95
filt_radius: 3
exp_name: experiment_1
"""

with open(config_yaml_path, 'w') as f:
    f.write(config_content)

ARGS = {
    'directory_path': temp_dir
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
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, 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
extract_epoch_from_filename Ran
ran; our contract draft was wrong, not the code
le-trungnguyen/neurips2024-activationcompression/classification/read_result.py
code served (permissive licence) · get_code("9dc3b9035dfd2ea6")
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 os
import re
import tempfile
import shutil

# Create a temporary directory structure with checkpoint files
temp_dir = tempfile.mkdtemp()
checkpoints_dir = os.path.join(temp_dir, 'checkpoints')
os.makedirs(checkpoints_dir, exist_ok=True)

# Create some checkpoint files matching the pattern
checkpoint_files = [
    'epoch=5-val-acc=0.95.ckpt',
    'epoch=10-val-acc=0.97.ckpt',
    'epoch=3-val-acc=0.85.ckpt',
    'other_file.txt'
]

for filename in checkpoint_files:
    filepath = os.path.join(checkpoints_dir, filename)
    with open(filepath, 'w') as f:
        f.write('dummy content')

ARGS = {
    'directory_path': temp_dir
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
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 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_latest_log_version Ran
ran, honoured the contract we drafted
le-trungnguyen/neurips2024-activationcompression/segmentation/train_count_mem.py
code served (permissive licence) · get_code("b7426c5f5b80730c")
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 int, ndim 0, dtype family int, shape () 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 os
import tempfile
import shutil

# Create a temporary directory structure with version_ subdirectories
temp_dir = tempfile.mkdtemp()

# Create some version_ directories
os.makedirs(os.path.join(temp_dir, 'version_0'), exist_ok=True)
os.makedirs(os.path.join(temp_dir, 'version_1'), exist_ok=True)
os.makedirs(os.path.join(temp_dir, 'version_5'), exist_ok=True)
os.makedirs(os.path.join(temp_dir, 'other_dir'), exist_ok=True)
os.makedirs(os.path.join(temp_dir, 'version_abc'), exist_ok=True)

ARGS = {
    'path': temp_dir
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
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 int, ndim 0, dtype family int, shape () — shape () 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
get_moment_logger Ran
ran; our contract draft was wrong, not the code
le-trungnguyen/neurips2024-activationcompression/segmentation/train_count_mem.py
code served (permissive licence) · get_code("2ad6871047208f3f")
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 function. 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 torch.nn as nn

torch.manual_seed(0)

class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
        self.moment1 = {}
        self.moment2 = {}
        self.moment_step = {}
    
    def forward(self, x):
        return self.linear(x)

model = SimpleModel()
name = "test_param"

ARGS = {
    "model": model,
    "name": name
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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_val_acc_for_epoch Ran
ran; our contract draft was wrong, not the code
le-trungnguyen/neurips2024-activationcompression/classification/read_result.py
code served (permissive licence) · get_code("ad9a0f0d585745c8")
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 os
import tempfile
import shutil

# Create a temporary directory for the test
temp_dir = tempfile.mkdtemp()

# Create a val.log file with sample validation accuracy data
val_log_content = """0 0.85
1 0.87
2 0.89
3 0.91
4 0.92
"""

val_log_path = os.path.join(temp_dir, 'val.log')
with open(val_log_path, 'w') as f:
    f.write(val_log_content)

ARGS = {
    'directory_path': temp_dir,
    'epoch': 2
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
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 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

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 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 8 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

Internet of Things and Deep Learning are synergetically and exponentially growing industrial fields with a massive call for their unification into a common framework called Edge AI. While on-device inference is a well-explored topic in recent research, backpropagation remains an open challenge due to its prohibitive computational and memory costs compared to the extreme resource constraints of embedded devices. Drawing on tensor decomposition research, we tackle the main bottleneck of backpropagation, namely the memory footprint of activation map storage. We investigate and compare the effects of activation compression using Singular Value Decomposition and its tensor variant, High-Order Singular Value Decomposition. The application of low-order decomposition results in considerable memory savings while preserving the features essential for learning, and also offers theoretical guarantees to convergence. Experimental results obtained on main-stream architectures and tasks demonstrate Pareto-superiority over other state-of-the-art solutions, in terms of the trade-off between generalization and memory footprint.

For agents

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

get_harvested_code_for_paper("2411.06346")
get_code_for_paper("2411.06346")
have("2411.06346")

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

curl https://syntology.ai/api/ran/2411.06346.json

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

[![Syntology run record](https://syntology.ai/api/ran/2411.06346.svg)](https://syntology.ai/paper/2411.06346)

Connect an agent — have() is free.