SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2204.07655 · CVPR · 2022

Deep Unlearning via Randomized Conditionally Independent Hessians

Vikas Singh, Sathya Ravi, Ronak Mehta

arXiv · PDF · Open in the Atlas

Code that ran

6 of 15 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 4 have no contract verdict; 2 failures of our instrument; 9 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/2204.07655.json · badge: /api/ran/2204.07655.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
vsingh-group/LCODEC-deep-unlearning canonical (classifier label)
classifier basis not recorded on the link
6 ran · 9 no recorded run
FunctionStatusWhere it lives
imagedata_kwargs Ran
ran on a synthesized input
vsingh-group/LCODEC-deep-unlearning/scrub/deep-person-reid-master/projects/OSNet_AIN/default_config.py
code served (permissive licence) · get_code("561c35d94c883194")
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 dict. 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)

class Config:
    def __init__(self):
        self.data = DataConfig()
        self.use_gpu = False
        self.train = TrainConfig()
        self.test = TestConfig()
        self.sampler = SamplerConfig()
        self.cuhk03 = CUHK03Config()
        self.market1501 = Market1501Config()

class DataConfig:
    def __init__(self):
        self.root = '/tmp/data'
        self.sources = ['market1501', 'dukemtmc']
        self.targets = ['msmt17']
        self.height = 256
        self.width = 128
        self.transforms = ['random_flip', 'random_crop']
        self.norm_mean = [0.485, 0.456, 0.406]
        self.norm_std = [0.229, 0.224, 0.225]
        self.split_id = 0
        self.combineall = False
        self.workers = 4

class TrainConfig:
    def __init__(self):
        self.batch_size = 32

class TestConfig:
    def __init__(self):
        self.batch_size = 16

class SamplerConfig:
    def __init__(self):
        self.num_instances = 4
        self.train_sampler = 'triplet'

class CUHK03Config:
    def __init__(self):
        self.labeled_images = True
        self.classic_split = False

class Market1501Config:
    def __init__(self):
        self.use_500k_distractors = False

cfg = Config()

ARGS = {'cfg': cfg}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
inceptionv4 Ran
ran on a synthesized input
vsingh-group/LCODEC-deep-unlearning/scrub/deep-person-reid-master/torchreid/models/inceptionv4.py
code served (permissive licence) · get_code("2bcd3828691ac1d1")
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 InceptionV4. 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)

# Build a minimal InceptionV4-like model for testing
class MinimalInceptionV4(nn.Module):
    def __init__(self, num_classes=1000):
        super(MinimalInceptionV4, self).__init__()
        self.conv1 = nn.Conv2d(3, 32, kernel_size=3, stride=2, padding=1)
        self.bn1 = nn.BatchNorm2d(32)
        self.relu = nn.ReLU(inplace=True)
        self.fc = nn.Linear(32 * 75 * 75, num_classes)
    
    def forward(self, x):
        x = self.conv1(x)
        x = self.bn1(x)
        x = self.relu(x)
        x = x.view(x.size(0), -1)
        x = self.fc(x)
        return x

# Create a small test input tensor (batch=2, channels=3, height=299, width=299)
test_input = torch.randn(2, 3, 299, 299)

# Create model instance
model = MinimalInceptionV4(num_classes=1000)

# Set model to eval mode
model.eval()

# Create the ARGS dict with all parameters for inceptionv4 function
ARGS = {
    'pretrained': False,
    'num_classes': 1000,
    'in_channels': 3
}
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 InceptionV4 — a InceptionV4 output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a InceptionV4 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
mlfn Ran
ran; our contract draft was wrong, not the code
vsingh-group/LCODEC-deep-unlearning/scrub/deep-person-reid-master/torchreid/models/mlfn.py
code served (permissive licence) · get_code("7f228eb824b42db8")
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 MLFN. 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
from torch.nn import functional as F

torch.manual_seed(0)

ARGS = {
    'num_classes': 10,
    'loss': 'softmax',
    'groups': 32,
    'channels': [64, 256, 512, 1024, 2048],
    'embed_dim': 1024,
}
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 MLFN — a MLFN output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a MLFN 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
optimizer_kwargs Ran
ran on a synthesized input
vsingh-group/LCODEC-deep-unlearning/scrub/deep-person-reid-master/projects/OSNet_AIN/default_config.py
code served (permissive licence) · get_code("d3e1b4b879ea0786")
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 dict. 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
from types import SimpleNamespace

torch.manual_seed(0)

# Create a config object with all required nested attributes
cfg = SimpleNamespace(
    train=SimpleNamespace(
        optim='sgd',
        lr=0.01,
        weight_decay=1e-4,
        staged_lr=False,
        new_layers=['layer4'],
        base_lr_mult=0.1
    ),
    sgd=SimpleNamespace(
        momentum=0.9,
        dampening=0.0,
        nesterov=True
    ),
    rmsprop=SimpleNamespace(
        alpha=0.99
    ),
    adam=SimpleNamespace(
        beta1=0.9,
        beta2=0.999
    )
)

ARGS = {'cfg': cfg}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
osnet_nas Ran
ran; our contract draft was wrong, not the code
vsingh-group/LCODEC-deep-unlearning/scrub/deep-person-reid-master/projects/OSNet_AIN/osnet_search.py
code served (permissive licence) · get_code("3e4d009ac7f89ad8")
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 OSNet. 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
from torch import nn

torch.manual_seed(0)

# Build minimal valid input for the ChannelGate.forward() method
# The forward method uses:
# - self.global_avgpool(x)
# - self.fc1(x)
# - self.norm1 (optional LayerNorm)
# - self.relu(x)
# - self.fc2(x)
# - self.gate_activation(x)
# - returns x (and optionally gates if return_gates=True)

# Create a minimal input tensor: (batch, channels, height, width)
x = torch.randn(2, 64, 8, 8)

ARGS = {
    'x': x,
}
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 OSNet — a OSNet output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a OSNet 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
videodata_kwargs Ran
ran on a synthesized input
vsingh-group/LCODEC-deep-unlearning/scrub/deep-person-reid-master/projects/OSNet_AIN/default_config.py
code served (permissive licence) · get_code("0951e0d511f67f90")
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 dict. 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 MockTransform:
    def __call__(self, x):
        return x

class MockConfig:
    def __init__(self):
        self.data = MockDataConfig()
        self.train = MockTrainConfig()
        self.test = MockTestConfig()
        self.sampler = MockSamplerConfig()
        self.video = MockVideoConfig()
        self.use_gpu = False

class MockDataConfig:
    def __init__(self):
        self.root = "/tmp/data"
        self.sources = ["camera1", "camera2"]
        self.targets = ["action", "person"]
        self.height = 224
        self.width = 224
        self.transforms = MockTransform()
        self.norm_mean = [0.485, 0.456, 0.406]
        self.norm_std = [0.229, 0.224, 0.225]
        self.split_id = 0
        self.combineall = False
        self.workers = 4

class MockTrainConfig:
    def __init__(self):
        self.batch_size = 32

class MockTestConfig:
    def __init__(self):
        self.batch_size = 16

class MockSamplerConfig:
    def __init__(self):
        self.num_instances = 4
        self.train_sampler = "random"

class MockVideoConfig:
    def __init__(self):
        self.seq_len = 8
        self.sample_method = "uniform"

cfg = MockConfig()

ARGS = {
    "cfg": cfg
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
build_model Not yet run
unverified (harvested, no recorded run)
vsingh-group/LCODEC-deep-unlearning/scrub/deep-person-reid-master/projects/OSNet_AIN/osnet_child.py
code served (permissive licence) · get_code("80c5da68bda0fa8c")
How it ran

no invocation recorded.

build_model Not yet run
unverified (harvested, no recorded run)
vsingh-group/LCODEC-deep-unlearning/scrub/deep-person-reid-master/projects/OSNet_AIN/osnet_search.py
code served (permissive licence) · get_code("d81cb662094c9cbf")
How it ran

no invocation recorded.

densenet121 Not yet run
unverified (harvested, no recorded run)
vsingh-group/LCODEC-deep-unlearning/scrub/deep-person-reid-master/torchreid/models/densenet.py
code served (permissive licence) · get_code("8e96abe9674da973")
How it ran

no invocation recorded.

densenet169 Not yet run
unverified (harvested, no recorded run)
vsingh-group/LCODEC-deep-unlearning/scrub/deep-person-reid-master/torchreid/models/densenet.py
code served (permissive licence) · get_code("54d10942ab6a6650")
How it ran

no invocation recorded.

densenet201 Not yet run
unverified (harvested, no recorded run)
vsingh-group/LCODEC-deep-unlearning/scrub/deep-person-reid-master/torchreid/models/densenet.py
code served (permissive licence) · get_code("372747120f68c12f")
How it ran

no invocation recorded.

inceptionresnetv2 Not yet run
unverified (harvested, no recorded run)
vsingh-group/LCODEC-deep-unlearning/scrub/deep-person-reid-master/torchreid/models/inceptionresnetv2.py
code served (permissive licence) · get_code("968ff7bf292adbec")
How it ran

no invocation recorded.

mobilenetv2_x1_0 Not yet run
unverified (harvested, no recorded run)
vsingh-group/LCODEC-deep-unlearning/scrub/deep-person-reid-master/torchreid/models/mobilenetv2.py
code served (permissive licence) · get_code("113c7b54049663fa")
How it ran

no invocation recorded.

mobilenetv2_x1_4 Not yet run
unverified (harvested, no recorded run)
vsingh-group/LCODEC-deep-unlearning/scrub/deep-person-reid-master/torchreid/models/mobilenetv2.py
code served (permissive licence) · get_code("3faa700e370230b2")
How it ran

no invocation recorded.

osnet_ain_x1_0 Not yet run
unverified (harvested, no recorded run)
vsingh-group/LCODEC-deep-unlearning/scrub/deep-person-reid-master/projects/OSNet_AIN/osnet_child.py
code served (permissive licence) · get_code("f3256be9107fa785")
How it ran

no invocation recorded.

Also present in other papers' code

Of 15 samples lifted from this paper's linked repositories, 15 carry a function identity and 0 do not. Of the 15 distinct functions with an identity, 15 are also present in other papers whose code we hold (15 with at least one paper dated earlier, 0 the same month, 0 later) and 0 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.

FunctionAlso present inEarliest-datedOthers named
densenet121 5 other papers
5 dated earlier
1905.00953 (2019-05)
Omni-Scale Feature Learning for Person Re-Identification
1910.10093 (2019-10)
Torchreid: A Library for Deep Learning Person Re-Identification in Pytorch
2010.05435 (2020-10)
Top-DB-Net: Top DropBlock for Activation Enhancement in Person Re-Identification
and 2 more not named here
densenet169 5 other papers
5 dated earlier
1905.00953 (2019-05)
Omni-Scale Feature Learning for Person Re-Identification
1910.10093 (2019-10)
Torchreid: A Library for Deep Learning Person Re-Identification in Pytorch
2010.05435 (2020-10)
Top-DB-Net: Top DropBlock for Activation Enhancement in Person Re-Identification
and 2 more not named here
densenet201 5 other papers
5 dated earlier
1905.00953 (2019-05)
Omni-Scale Feature Learning for Person Re-Identification
1910.10093 (2019-10)
Torchreid: A Library for Deep Learning Person Re-Identification in Pytorch
2010.05435 (2020-10)
Top-DB-Net: Top DropBlock for Activation Enhancement in Person Re-Identification
and 2 more not named here
inceptionresnetv2 5 other papers
5 dated earlier
1905.00953 (2019-05)
Omni-Scale Feature Learning for Person Re-Identification
1910.10093 (2019-10)
Torchreid: A Library for Deep Learning Person Re-Identification in Pytorch
2010.05435 (2020-10)
Top-DB-Net: Top DropBlock for Activation Enhancement in Person Re-Identification
and 2 more not named here
inceptionv4 5 other papers
5 dated earlier
1905.00953 (2019-05)
Omni-Scale Feature Learning for Person Re-Identification
1910.10093 (2019-10)
Torchreid: A Library for Deep Learning Person Re-Identification in Pytorch
2010.05435 (2020-10)
Top-DB-Net: Top DropBlock for Activation Enhancement in Person Re-Identification
and 2 more not named here
mlfn 5 other papers
5 dated earlier
1905.00953 (2019-05)
Omni-Scale Feature Learning for Person Re-Identification
1910.10093 (2019-10)
Torchreid: A Library for Deep Learning Person Re-Identification in Pytorch
2010.05435 (2020-10)
Top-DB-Net: Top DropBlock for Activation Enhancement in Person Re-Identification
and 2 more not named here
mobilenetv2_x1_0 4 other papers
4 dated earlier
1905.00953 (2019-05)
Omni-Scale Feature Learning for Person Re-Identification
1910.10093 (2019-10)
Torchreid: A Library for Deep Learning Person Re-Identification in Pytorch
2010.05435 (2020-10)
Top-DB-Net: Top DropBlock for Activation Enhancement in Person Re-Identification
and 1 more not named here
mobilenetv2_x1_4 4 other papers
4 dated earlier
1905.00953 (2019-05)
Omni-Scale Feature Learning for Person Re-Identification
1910.10093 (2019-10)
Torchreid: A Library for Deep Learning Person Re-Identification in Pytorch
2010.05435 (2020-10)
Top-DB-Net: Top DropBlock for Activation Enhancement in Person Re-Identification
and 1 more not named here
optimizer_kwargs 4 other papers
4 dated earlier
1910.10093 (2019-10)
Torchreid: A Library for Deep Learning Person Re-Identification in Pytorch
2010.05435 (2020-10)
Top-DB-Net: Top DropBlock for Activation Enhancement in Person Re-Identification
Zhao_PhD_Learning_Learning_With_Pompeiu-Hausdorff_Distances_for_Video-Based_Vehicle_Re-Identification_CVPR_2021_paper (2021 (year only))
PhD Learning: Learning with Pompeiu-hausdorff Distances for Video-based Vehicle Re-Identification
and 1 more not named here
videodata_kwargs 4 other papers
3 dated earlier, 1 same year, order not determinable
1910.10093 (2019-10)
Torchreid: A Library for Deep Learning Person Re-Identification in Pytorch
2010.05435 (2020-10)
Top-DB-Net: Top DropBlock for Activation Enhancement in Person Re-Identification
Zhao_PhD_Learning_Learning_With_Pompeiu-Hausdorff_Distances_for_Video-Based_Vehicle_Re-Identification_CVPR_2021_paper (2021 (year only))
PhD Learning: Learning with Pompeiu-hausdorff Distances for Video-based Vehicle Re-Identification
and 1 more not named here
imagedata_kwargs 3 other papers
3 dated earlier
1910.10093 (2019-10)
Torchreid: A Library for Deep Learning Person Re-Identification in Pytorch
2010.05435 (2020-10)
Top-DB-Net: Top DropBlock for Activation Enhancement in Person Re-Identification
Zhao_PhD_Learning_Learning_With_Pompeiu-Hausdorff_Distances_for_Video-Based_Vehicle_Re-Identification_CVPR_2021_paper (2021 (year only))
PhD Learning: Learning with Pompeiu-hausdorff Distances for Video-based Vehicle Re-Identification
build_model 2 other papers
2 dated earlier
1910.10093 (2019-10)
Torchreid: A Library for Deep Learning Person Re-Identification in Pytorch
Zhao_PhD_Learning_Learning_With_Pompeiu-Hausdorff_Distances_for_Video-Based_Vehicle_Re-Identification_CVPR_2021_paper (2021 (year only))
PhD Learning: Learning with Pompeiu-hausdorff Distances for Video-based Vehicle Re-Identification
build_model 2 other papers
2 dated earlier
1910.10093 (2019-10)
Torchreid: A Library for Deep Learning Person Re-Identification in Pytorch
Zhao_PhD_Learning_Learning_With_Pompeiu-Hausdorff_Distances_for_Video-Based_Vehicle_Re-Identification_CVPR_2021_paper (2021 (year only))
PhD Learning: Learning with Pompeiu-hausdorff Distances for Video-based Vehicle Re-Identification
osnet_ain_x1_0 2 other papers
2 dated earlier
1910.10093 (2019-10)
Torchreid: A Library for Deep Learning Person Re-Identification in Pytorch
Zhao_PhD_Learning_Learning_With_Pompeiu-Hausdorff_Distances_for_Video-Based_Vehicle_Re-Identification_CVPR_2021_paper (2021 (year only))
PhD Learning: Learning with Pompeiu-hausdorff Distances for Video-based Vehicle Re-Identification
osnet_nas 2 other papers
2 dated earlier
1910.10093 (2019-10)
Torchreid: A Library for Deep Learning Person Re-Identification in Pytorch
Zhao_PhD_Learning_Learning_With_Pompeiu-Hausdorff_Distances_for_Video-Based_Vehicle_Re-Identification_CVPR_2021_paper (2021 (year only))
PhD Learning: Learning with Pompeiu-hausdorff Distances for Video-based Vehicle Re-Identification

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

Recent legislation has led to interest in machine unlearning, i.e., removing specific training samples from a predictive model as if they never existed in the training dataset. Unlearning may also be required due to corrupted/adversarial data or simply a user's updated privacy requirement. For models which require no training (k-NN), simply deleting the closest original sample can be effective. But this idea is inapplicable to models which learn richer representations. Recent ideas leveraging optimization-based updates scale poorly with the model dimension d, due to inverting the Hessian of the loss function. We use a variant of a new conditional independence coefficient, L-CODEC, to identify a subset of the model parameters with the most semantic overlap on an individual sample level. Our approach completely avoids the need to invert a (possibly) huge matrix. By utilizing a Markov blanket selection, we premise that L-CODEC is also suitable for deep unlearning, as well as other applications in vision. Compared to alternatives, L-CODEC makes approximate unlearning possible in settings that would otherwise be infeasible, including vision models used for face recognition, person reidentification and NLP models that may require unlearning samples identified for exclusion. Code is available at https://github.com/vsingh-group/LCODECdeep-unlearning

For agents

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

get_harvested_code_for_paper("2204.07655")
get_code_for_paper("2204.07655")
have("2204.07655")

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

curl https://syntology.ai/api/ran/2204.07655.json

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

[![Syntology run record](https://syntology.ai/api/ran/2204.07655.svg)](https://syntology.ai/paper/2204.07655)

Connect an agent — have() is free.