SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2602.08913 · 2026

GEMSS: A Variational Method for Discovering Multiple Sparse Solutions in Classification and Regression Problems

Kateřina Henclová, Václav Šmídl

arXiv · PDF · Open in the Atlas

Code that ran

16 of 21 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; 13 failures of our instrument; 5 no recorded run), measured 2026-11-22; 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/2602.08913.json · badge: /api/ran/2602.08913.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
kat-er-ina/gemss_testing canonical (classifier label)
classifier basis: census
12 ran · 1 no recorded run
kat-er-ina/gemss canonical (classifier label)
classifier basis: census
4 ran · 4 no recorded run
FunctionStatusWhere it lives
calculate_metrics Ran
ran; our contract draft was wrong, not the code
kat-er-ina/gemss_testing/algorithm_comparison/src/evaluation.py
pointer only (licence: NONE) · get_code("5e372b1774b9c227")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, kat-er-ina/gemss_testing 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.

calculate_structural_metrics Ran
ran; our contract draft was wrong, not the code
kat-er-ina/gemss_testing/algorithm_comparison/src/evaluation.py
pointer only (licence: NONE) · get_code("b7ba0e35ec109577")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, kat-er-ina/gemss_testing 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.

case2set Ran
ran; our contract draft was wrong, not the code
kat-er-ina/gemss/gemss/experiment_assessment/case_analysis.py
code served (permissive licence) · get_code("e3acd7ac32186cfe")
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

# The function case2set only takes one parameter: case_id (int)
# It maps a case ID to its corresponding case set by checking membership in ranges
# We need to provide an integer case_id that exists in one of the ranges defined in CASE_SET_RANGES

CASE_SET_RANGES = {
    'baseline': [1, 2, 3, 14, 15, 16],
    'scalability': range(4, 11),
    'samplerich': range(11, 14),
    'adversity': range(17, 24),
    'jaccard': range(24, 29),
    'unbalanced': range(29, 32),
    'reg_baseline': range(32, 34),
    'reg_scalability': range(34, 40),
    'reg_adversity': range(40, 43),
    'reg_vs_class': range(43, 48)
}

def case2set(case_id: int) -> str:
    """
    Maps a case ID to its corresponding case set.

    Parameters
    ----------
    case_id : int
        The case ID to map.

    Returns
    -------
    str
        The name of the case set the case ID belongs to.
    """
    for case_set, case_range in CASE_SET_RANGES.items():
        if case_id in case_range:
            return case_set
    raise ValueError(f'Case ID {case_id} not found in any case set.')


# Valid case_id that exists in the ranges (e.g., 1 is in baseline)
ARGS = {
    'case_id': 5
}
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
cluster_supports Ran
ran; our contract draft was wrong, not the code
kat-er-ina/gemss_testing/algorithm_comparison/diagnostics/ensemble_parity.py
pointer only (licence: NONE) · get_code("611a7bb17d4ce09a")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, kat-er-ina/gemss_testing 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.

cluster_to_m Ran
ran; our contract draft was wrong, not the code
kat-er-ina/gemss_testing/algorithm_comparison/diagnostics/bb_ssl_probe.py
pointer only (licence: NONE) · get_code("d81f7dd887ede59c")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, kat-er-ina/gemss_testing 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.

concatenate_cases Ran
ran; our contract draft was wrong, not the code
kat-er-ina/gemss/gemss/experiment_assessment/case_analysis.py
code served (permissive licence) · get_code("d4ad6f14626c11d6")
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 DataFrame. 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 pandas as pd
import numpy as np

# Create sample dataframes for different cases
df_case_1 = pd.DataFrame({
    'feature_1': [1.0, 2.0],
    'feature_2': ['a', 'b'],
    'value': [10, 20]
})

df_case_2 = pd.DataFrame({
    'feature_1': [3.0, 4.0],
    'feature_2': ['c', 'd'],
    'value': [30, 40]
})

df_case_3 = pd.DataFrame({
    'feature_1': [5.0],
    'feature_2': ['e'],
    'value': [50]
})

# Construct the dictionary of case dataframes
df_cases = {
    1: df_case_1,
    2: df_case_2,
    3: df_case_3
}

ARGS = {
    'df_cases': df_cases
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DataFrame — a DataFrame output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataFrame output the recorded ndim 2 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
fnum Ran
ran, honoured the contract we drafted
kat-er-ina/gemss_testing/algorithm_comparison/diagnostics/stats_agg.py
pointer only (licence: NONE) · get_code("374885459d29217c")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, kat-er-ina/gemss_testing 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.

generate_multi_solution_data Ran
ran; our contract draft was wrong, not the code
kat-er-ina/gemss/gemss/data_handling/generate_artificial_dataset.py
code served (permissive licence) · get_code("54f97ea347da57c8")
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. 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 numpy as np
import pandas as pd

ARGS = {
    'n_samples': 50,
    'n_features': 8,
    'n_solutions': 3,
    'sparsity': 2,
    'noise_data_std': 0.01,
    'binarize': True,
    'binary_response_ratio': 0.5,
    'random_seed': 42,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas — 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
generate_overlapping_dataset Ran
ran; our contract draft was wrong, not the code
kat-er-ina/gemss_testing/algorithm_comparison/src/hard_data_factory.py
pointer only (licence: NONE) · get_code("e1b61e8d1483509f")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, kat-er-ina/gemss_testing 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.

get_feature_name_mapping Ran
ran; our contract draft was wrong, not the code
kat-er-ina/gemss/gemss/data_handling/data_processing.py
code served (permissive licence) · get_code("eab9128873931edc")
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 pandas as pd

# Create a simple DataFrame with a few columns
df = pd.DataFrame({
    'age': [25, 30, 35],
    'salary': [50000, 60000, 70000],
    'experience': [2, 5, 8],
    'department': ['IT', 'HR', 'Finance']
})

ARGS = {
    'df': df
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pandas — 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
grid Ran
ran; our contract draft was wrong, not the code
kat-er-ina/gemss_testing/algorithm_comparison/diagnostics/tuned_compare.py
pointer only (licence: NONE) · get_code("717e32343d81591f")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, kat-er-ina/gemss_testing 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.

load Ran
ran; our contract draft was wrong, not the code
kat-er-ina/gemss_testing/algorithm_comparison/diagnostics/stats_agg.py
pointer only (licence: NONE) · get_code("0b1bf82b4f68387b")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, kat-er-ina/gemss_testing 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.

per_seed_at_best Ran
ran; our contract draft was wrong, not the code
kat-er-ina/gemss_testing/algorithm_comparison/diagnostics/stats_agg.py
pointer only (licence: NONE) · get_code("51ef4e782dbd034a")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, kat-er-ina/gemss_testing 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.

perplexity Ran
ran, honoured the contract we drafted
kat-er-ina/gemss_testing/algorithm_comparison/diagnostics/rq5_count.py
pointer only (licence: NONE) · get_code("d0aac9ca6073644b")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, kat-er-ina/gemss_testing 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.

predictive_quality Ran
ran; our contract draft was wrong, not the code
kat-er-ina/gemss_testing/algorithm_comparison/src/evaluation.py
pointer only (licence: NONE) · get_code("27e46af6b8cf6e9c")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, kat-er-ina/gemss_testing 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.

support_recovery_check Ran
ran on a synthesized input
kat-er-ina/gemss_testing/algorithm_comparison/src/hard_data_factory.py
pointer only (licence: NONE) · get_code("6987276b3c379eef")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, kat-er-ina/gemss_testing 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.

get_current_config Not yet run
unverified (harvested, no recorded run)
kat-er-ina/gemss/gemss/config/config.py
code served (permissive licence) · get_code("f431064ae42a70ec")
How it ran

no invocation recorded.

get_df_cases Not yet run
unverified (harvested, no recorded run)
kat-er-ina/gemss/gemss/experiment_assessment/case_analysis.py
code served (permissive licence) · get_code("19a01f4f22f9d6f9")
How it ran

no invocation recorded.

get_params_by_category Not yet run
unverified (harvested, no recorded run)
kat-er-ina/gemss/gemss/config/config.py
code served (permissive licence) · get_code("d74e72b488187ee7")
How it ran

no invocation recorded.

get_recommendation_message Not yet run
unverified (harvested, no recorded run)
kat-er-ina/gemss/gemss/diagnostics/recommendation_messages.py
code served (permissive licence) · get_code("e66085c8d8e9ee07")
How it ran

no invocation recorded.

run_bbssl Not yet run
unverified (harvested, no recorded run)
kat-er-ina/gemss_testing/algorithm_comparison/diagnostics/bb_ssl_probe.py
pointer only (licence: NONE) · get_code("4452d07a638099f3")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, kat-er-ina/gemss_testing 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 21 samples lifted from this paper's linked repositories, 21 carry a function identity and 0 do not. Of the 21 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 21 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

In underdetermined regression and classification problems, multiple feature subsets often yield equivalent predictive performance. In applied settings, especially with n ≪ p, high dimension or collinearities, it is valuable to provide a domain expert with a menu of statistically plausible explanations, rather than one arbitrary solution. This creates the need for appropriate methods. We present Gaussian Ensemble for Multiple Sparse Solutions (GEMSS), a method that uses a single variational mixture to approximate the corresponding multimodal posterior. Its evidence lower bound contains a built-in repulsion between the mixture's components, enabling the model to simultaneously produce several distinct sparse solutions. We evaluate GEMSS on a novel, reusable benchmark. The ground-truth solution set and its structure are known by construction and set-level recovery metrics are evaluated. GEMSS consistently outperforms dedicated multiplicity methods (Enumeration LASSO, ALFESE), two strong sampling baselines that approximate the same posterior (Randomized-LASSO ensemble, BB-SSL), and naive iterative masking. As solutions' overlap increases, the gap widens and additional ensemble restarts cannot close it. Only ALFESE proves competitive. Further, GEMSS is validated on real-world datasets, producing multiple distinct and highly predictive solutions: the practical goal that existing methods struggle to meet. The open-source Python package gemss is available (https://github.com/kat-er-ina/gemss) and democratized through a free online application at https://huggingface.co/spaces/ kat-er-ina/gemss.

For agents

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

get_harvested_code_for_paper("2602.08913")
get_code_for_paper("2602.08913")
have("2602.08913")

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

curl https://syntology.ai/api/ran/2602.08913.json

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

[![Syntology run record](https://syntology.ai/api/ran/2602.08913.svg)](https://syntology.ai/paper/2602.08913)

Connect an agent — have() is free.