SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2205.08096 · 2022

Can Bad Teaching Induce Forgetting? Unlearning in Deep Networks using an Incompetent Teacher

arXiv · PDF · Open in the Atlas

Code that ran

8 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; 2 honoured the contract we drafted, 0 violated it and 1 has no contract verdict; 5 failures of our instrument; 1 no recorded run), measured 2026-11-21; 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/2205.08096.json · badge: /api/ran/2205.08096.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
vikram2000b/bad-teaching-unlearning canonical (classifier label)
classifier basis not recorded on the link
8 ran · 1 no recorded run
FunctionStatusWhere it lives
JSDiv Ran
ran; our contract draft was wrong, not the code
vikram2000b/bad-teaching-unlearning/metrics.py
pointer only (licence: MIT); the stored copy get_code serves, from jwf40/zeroshot-unlearning-at-scale, is pointer only (licence: NONE) · get_code("65e6e724245e6071")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from jwf40/zeroshot-unlearning-at-scale, and GitHub detected no licence in that repository 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.

UnLearningScore Ran
ran, honoured the contract we drafted
vikram2000b/bad-teaching-unlearning/metrics.py
pointer only (licence: MIT); the stored copy get_code serves, from jwf40/zeroshot-unlearning-at-scale, is pointer only (licence: NONE) · get_code("e982685418b19915")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from jwf40/zeroshot-unlearning-at-scale, and GitHub detected no licence in that repository 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.

UnlearnerLoss Ran
ran; our contract draft was wrong, not the code
vikram2000b/bad-teaching-unlearning/unlearn.py
pointer only (licence: MIT); the stored copy get_code serves, from jwf40/information-theoretic-unlearning, is pointer only (licence: NONE) · get_code("35a220c6eeea45d4")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from jwf40/information-theoretic-unlearning, and GitHub detected no licence in that repository 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.

accuracy Ran
ran; our contract draft was wrong, not the code
vikram2000b/bad-teaching-unlearning/utils.py
pointer only (licence: MIT); the stored copy get_code serves, from OngWinKent/Federated-Feature-Unlearning, is pointer only (licence: NOASSERTION) · get_code("e8dbb156b1109cfa")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from OngWinKent/Federated-Feature-Unlearning, and GitHub could not classify the licence of that repository (NOASSERTION). 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.

entropy Ran
ran; our contract draft was wrong, not the code
vikram2000b/bad-teaching-unlearning/metrics.py
pointer only (licence: MIT); the stored copy get_code serves, from jwf40/zeroshot-unlearning-at-scale, is pointer only (licence: NONE) · get_code("4a12bd3434683bbc")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from jwf40/zeroshot-unlearning-at-scale, and GitHub detected no licence in that repository 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.

training_step Ran
ran on a synthesized input
vikram2000b/bad-teaching-unlearning/utils.py
code served (permissive licence) · get_code("cc6e4e27f3ee438c")
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 Tensor, ndim 0, dtype family float, 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
import torch.nn as nn
from torch.nn import functional as F

torch.manual_seed(0)

# Define a small model that takes images and outputs logits for classification
class SimpleModel(nn.Module):
    def __init__(self, num_classes=5):
        super().__init__()
        self.fc = nn.Linear(4, num_classes)
    
    def forward(self, x):
        # Flatten if needed and pass through linear layer
        x = x.view(x.size(0), -1)
        return self.fc(x)

# Create model
model = SimpleModel(num_classes=5)

# Create batch: (images, labels, clabels)
# images: shape (2, 4) - batch of 2, feature dim 4
# labels: not used in function, but included in batch tuple
# clabels: shape (2,) - class labels for cross_entropy
images = torch.randn(2, 4)
labels = torch.tensor([0, 1])
clabels = torch.tensor([1, 3])
batch = (images, labels, clabels)

# Device
device = 'cpu'

ARGS = {
    'model': model,
    'batch': batch,
    'device': device
}
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 Tensor, ndim 0, dtype family float, 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
unlearning_step Ran
ran, honoured the contract we drafted
vikram2000b/bad-teaching-unlearning/unlearn.py
pointer only (licence: MIT); the stored copy get_code serves, from jwf40/information-theoretic-unlearning, is pointer only (licence: NONE) · get_code("7920183b7b2256da")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from jwf40/information-theoretic-unlearning, and GitHub detected no licence in that repository 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.

validation_step Ran
ran; our contract draft was wrong, not the code
vikram2000b/bad-teaching-unlearning/utils.py
code served (permissive licence) · get_code("ea77dadbbf5a920e")
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 torch.nn import functional as F

torch.manual_seed(0)

# Define a minimal model that takes images and outputs logits
class SimpleModel(nn.Module):
    def __init__(self, num_classes=5):
        super().__init__()
        self.fc = nn.Linear(4, num_classes)
    
    def forward(self, x):
        # Expect input shape (batch_size, 4)
        return self.fc(x)

# Create instances
model = SimpleModel(num_classes=5)
device = 'cpu'

# Create a batch: (images, labels, clabels)
# images: (batch_size=2, feature_dim=4)
# labels: not used in validation_step, but included in batch
# clabels: (batch_size=2,) with class indices
batch = (
    torch.randn(2, 4),           # images
    torch.tensor([0, 1]),         # labels (unused)
    torch.tensor([1, 3])          # clabels (classification labels)
)

ARGS = {
    'model': model,
    'batch': batch,
    'device': device
}
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
fit_one_unlearning_cycle Not yet run
unverified (harvested, no recorded run)
vikram2000b/bad-teaching-unlearning/unlearn.py
pointer only (licence: MIT); the stored copy get_code serves, from jwf40/information-theoretic-unlearning, is pointer only (licence: NONE) · get_code("015561a0364e1ddb")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from jwf40/information-theoretic-unlearning, and GitHub detected no licence in that repository 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. 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, 8 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 8 later) and 1 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
entropy 4 other papers
1 dated earlier, 3 later
2201.05629 (2022-01)
Zero-Shot Machine Unlearning
2308.07707 (2023-08)
Fast Machine Unlearning Without Retraining Through Selective Synaptic Dampening
2310.18574 (2023-10)
Breaking the Trilemma of Privacy, Utility, Efficiency via Controllable Machine Unlearning
and 1 more not named here
JSDiv 2 other papers
2 later
2308.07707 (2023-08)
Fast Machine Unlearning Without Retraining Through Selective Synaptic Dampening
2402.01401 (2024-02)
An Information Theoretic Approach to Machine Unlearning
UnLearningScore 2 other papers
2 later
2308.07707 (2023-08)
Fast Machine Unlearning Without Retraining Through Selective Synaptic Dampening
2402.01401 (2024-02)
An Information Theoretic Approach to Machine Unlearning
fit_one_unlearning_cycle 2 other papers
2 later
2402.01401 (2024-02)
An Information Theoretic Approach to Machine Unlearning
2403.01267 (2024-03)
Dissecting Language Models: Machine Unlearning via Selective Pruning
unlearning_step 2 other papers
2 later
2402.01401 (2024-02)
An Information Theoretic Approach to Machine Unlearning
2403.01267 (2024-03)
Dissecting Language Models: Machine Unlearning via Selective Pruning
UnlearnerLoss 1 other paper
1 later
2402.01401 (2024-02)
An Information Theoretic Approach to Machine Unlearning
accuracy 1 other paper
1 later
2405.17462 (2024-05)
Ferrari: Federated Feature Unlearning via Optimizing Feature Sensitivity
validation_step 1 other paper
1 later
2405.17462 (2024-05)
Ferrari: Federated Feature Unlearning via Optimizing Feature Sensitivity

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 2026-11-21.

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 unlearning has become an important area of research due to an increasing need for machine learning (ML) applications to comply with the emerging data privacy regulations. It facilitates the provision for removal of certain set or class of data from an already trained ML model without requiring retraining from scratch. Recently, several efforts have been put in to make unlearning to be effective and efficient. We propose a novel machine unlearning method by exploring the utility of competent and incompetent teachers in a student-teacher framework to induce forgetfulness. The knowledge from the competent and incompetent teachers is selectively transferred to the student to obtain a model that doesn't contain any information about the forget data. We experimentally show that this method generalizes well, is fast and effective. Furthermore, we introduce the zero retrain forgetting (ZRF) metric to evaluate any unlearning method. Unlike the existing unlearning metrics, the ZRF score does not depend on the availability of the expensive retrained model. This makes it useful for analysis of the unlearned model after deployment as well. We present results of experiments conducted for random subset forgetting and class forgetting on various deep networks and across different application domains.~Source code is at: https://github.com/vikram2000b/bad-teaching-unlearning

For agents

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

get_harvested_code_for_paper("2205.08096")
get_code_for_paper("2205.08096")
have("2205.08096")

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

curl https://syntology.ai/api/ran/2205.08096.json

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

[![Syntology run record](https://syntology.ai/api/ran/2205.08096.svg)](https://syntology.ai/paper/2205.08096)

Connect an agent — have() is free.