SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2201.12919 · ICML · 2022

Provable Domain Generalization via Invariant-Feature Subspace Recovery

Bo Li, Han Zhao, Haoxiang Wang, Haozhe Si

arXiv · PDF · Open in the Atlas

Code that ran

2 of 3 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (2 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 2 have no contract verdict; 0 failures of our instrument; 1 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/2201.12919.json · badge: /api/ran/2201.12919.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
haoxiang-wang/isr alias (classifier label)
classifier basis: classifier_2025
2 ran · 1 no recorded run
FunctionStatusWhere it lives
ERM Ran
ran on a synthesized input · constructed an object; no result was computed
haoxiang-wang/isr/linear_unit_tests/isr.py
code served (permissive licence) · get_code("ea8e9b41e123be6a")
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 an instance of its own class, ERM: the run constructed the object and did not call it, so no result was computed. 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 random

torch.manual_seed(0)
random.seed(0)

ARGS = {
    'in_features': 4,
    'out_features': 3,
    'task': 'regression',
    'hparams': 'default',
    'regression': False,
    'num_iterations': 10
}
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 ERM — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a ERM 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
ISR Ran
ran on a synthesized input · constructed an object; no result was computed
haoxiang-wang/isr/linear_unit_tests/isr.py
code served (permissive licence) · get_code("c55e18a0dc0b725a")
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 an instance of its own class, ISR: the run constructed the object and did not call it, so no result was computed. 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
from sklearn.linear_model import LogisticRegression, Ridge
import scipy
import random
import json

torch.manual_seed(0)
np.random.seed(0)

# Create sample environments: list of (data, labels) tuples
# Each environment has 20 samples, 5 features
env1_data = np.random.randn(20, 5).astype(np.float32)
env1_labels = np.random.randint(0, 2, 20)
env2_data = np.random.randn(20, 5).astype(np.float32)
env2_labels = np.random.randint(0, 2, 20)
env3_data = np.random.randn(20, 5).astype(np.float32)
env3_labels = np.random.randint(0, 2, 20)

envs = [
    (env1_data, env1_labels),
    (env2_data, env2_labels),
    (env3_data, env3_labels),
]

ARGS = {
    'dim_inv': 3,
    'fit_method': 'cov',
    'l2_reg': 0.01,
    'verbose': False,
    'regression': False,
    'spu_proj': False,
    'logistic_regression': False,
    'hparams': None,
    'num_iterations': 1000,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy, sklearn, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ISR — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a ISR 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
Model Not yet run
unverified (harvested, no recorded run)
haoxiang-wang/isr/linear_unit_tests/isr.py
code served (permissive licence) · get_code("f84fe481344293cd")
How it ran

no invocation recorded.

Also present in other papers' code

Of 3 samples lifted from this paper's linked repositories, 3 carry a function identity and 0 do not. Of the 3 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 3 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

Domain generalization asks for models trained over a set of training environments to perform well in unseen test environments. Recently, a series of algorithms such as Invariant Risk Minimization (IRM) has been proposed for domain generalization. However, Rosenfeld et al. (2021) shows that in a simple linear data model, even if non-convexity issues are ignored, IRM and its extensions cannot generalize to unseen environments with less than d s `1 training environments, where d s is the dimension of the spuriousfeature subspace. In this paper, we propose to achieve domain generalization with Invariantfeature Subspace Recovery (ISR). Our first algorithm, ISR-Mean, can identify the subspace spanned by invariant features from the first-order moments of the class-conditional distributions, and achieve provable domain generalization with d s `1 training environments under the data model of Rosenfeld et al. (2021). Our second algorithm, ISR-Cov, further reduces the required number of training environments to Op1q using the information of second-order moments. Notably, unlike IRM, our algorithms bypass non-convexity issues and enjoy global convergence guarantees. Empirically, our ISRs can obtain superior performance compared with IRM on synthetic benchmarks. In addition, on three real-world image and text datasets, we show that both ISRs can be used as simple yet effective post-processing methods to improve the worst-case accuracy of (pre-)trained models against spurious correlations and group shifts. The code is released at https: //github.com/Haoxiang-Wang/ISR.

For agents

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

get_harvested_code_for_paper("2201.12919")
get_code_for_paper("2201.12919")
have("2201.12919")

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

curl https://syntology.ai/api/ran/2201.12919.json

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

[![Syntology run record](https://syntology.ai/api/ran/2201.12919.svg)](https://syntology.ai/paper/2201.12919)

Connect an agent — have() is free.