SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1609.04747 · 2016

An overview of gradient descent optimization algorithms

arXiv · PDF · Open in the Atlas

Code that ran

5 of 7 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 5 have no contract verdict; 0 failures of our instrument; 2 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/1609.04747.json · badge: /api/ran/1609.04747.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
biocore/mmvec community (classifier label)
classifier basis not recorded on the link
4 ran · 0 no recorded run
biocore/songbird community (classifier label)
classifier basis not recorded on the link
1 ran · 1 no recorded run
mortonjt/songbird community (classifier label)
classifier basis not recorded on the link
0 ran · 1 no recorded run
FunctionStatusWhere it lives
paired_heatmaps Ran
ran on a synthesized input
biocore/mmvec/mmvec/heatmap.py
code served (permissive licence) · get_code("b2febb39c1c1afcc")
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 pandas as pd
import numpy as np

# Create a small stand-in class for biom.Table
class BiomTable:
    def __init__(self, data, sample_ids, observation_ids):
        self.data = data
        self.sample_ids = sample_ids
        self.observation_ids = observation_ids
    
    def ids(self, axis):
        if axis == 'observation':
            return self.observation_ids
        elif axis == 'sample':
            return self.sample_ids
    
    def to_dataframe(self):
        df = pd.DataFrame(self.data, index=self.observation_ids, columns=self.sample_ids)
        return df

# Create sample data
np.random.seed(0)

# Microbe feature IDs
microbe_ids = ['microbe_0', 'microbe_1', 'microbe_2']
# Sample IDs
sample_ids = ['sample_0', 'sample_1']
# Metabolite IDs
metabolite_ids = ['metabolite_0', 'metabolite_1', 'metabolite_2', 'metabolite_3']

# Create microbes_table (features x samples)
microbes_data = np.random.rand(3, 2) * 100
microbes_table = BiomTable(microbes_data, sample_ids, microbe_ids)

# Create metabolites_table (features x samples)
metabolites_data = np.random.rand(4, 2) * 50
metabolites_table = BiomTable(metabolites_data, sample_ids, metabolite_ids)

# Create ranks DataFrame (microbes x metabolites) - conditional probabilities
ranks = pd.DataFrame(
    np.random.rand(3, 4),
    index=microbe_ids,
    columns=metabolite_ids
)

# Create microbe_metadata Series with semicolon-delimited taxonomy strings
microbe_metadata = pd.Series(
    {
        'microbe_0': 'Bacteria;Firmicutes;Bacilli;Bacillales',
        'microbe_1': 'Bacteria;Proteobacteria;Gammaproteobacteria;Enterobacterales',
        'microbe_2': 'Bacteria;Actinobacteria;Actinobacteria;Actinomycetales',
    }
)

ARGS = {
    'ranks': ranks,
    'microbes_table': microbes_table,
    'metabolites_table': metabolites_table,
    'microbe_metadata': microbe_metadata,
    'features': None,
    'top_k_microbes': 2,
    'top_k_metabolites': 3,
    'keep_top_samples': True,
    'level': 2,
    'normalize': 'log10',
    'color_palette': 'magma'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
matplotlib, numpy, pandas, seaborn — 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
rank_hits Ran
ran on a synthesized input
biocore/mmvec/mmvec/util.py
code served (permissive licence) · get_code("1a5ac4a378e02613")
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 numpy as np
import pandas as pd

# Create a small rank matrix (DataFrame with numerical values)
# Rows represent source nodes, columns represent destination nodes
# Values are ranks/conditional probabilities
np.random.seed(0)
ranks_data = np.random.rand(3, 4)  # 3 rows, 4 columns with distinct dimensions
ranks = pd.DataFrame(
    ranks_data,
    index=['node_a', 'node_b', 'node_c'],
    columns=['dest_1', 'dest_2', 'dest_3', 'dest_4']
)

ARGS = {
    'ranks': ranks,
    'k': 2,
    'pos': True
}
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
ranks_heatmap Ran
ran on a synthesized input
biocore/mmvec/mmvec/heatmap.py
code served (permissive licence) · get_code("75ccd9ba6246b881")
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 ClusterGrid. 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
import seaborn as sns
import matplotlib.pyplot as plt

# Create a small DataFrame for ranks (microbes x metabolites)
# Using distinct dimensions: 5 microbes, 7 metabolites
np.random.seed(0)
ranks_data = np.random.randn(5, 7) * 0.5
ranks = pd.DataFrame(
    ranks_data,
    index=[f'microbe_{i}' for i in range(5)],
    columns=[f'metabolite_{j}' for j in range(7)]
)

# Create microbe metadata Series
microbe_metadata = pd.Series(
    ['Genus_A;Species_X;Strain_1', 'Genus_B;Species_Y;Strain_2', 
     'Genus_A;Species_X;Strain_3', 'Genus_C;Species_Z;Strain_4',
     'Genus_B;Species_Y;Strain_5'],
    index=[f'microbe_{i}' for i in range(5)],
    name='taxonomy'
)

# Create metabolite metadata Series
metabolite_metadata = pd.Series(
    ['TypeA', 'TypeB', 'TypeA', 'TypeC', 'TypeB', 'TypeA', 'TypeC'],
    index=[f'metabolite_{j}' for j in range(7)],
    name='metabolite_type'
)

ARGS = {
    'ranks': ranks,
    'microbe_metadata': microbe_metadata,
    'metabolite_metadata': metabolite_metadata,
    'method': 'average',
    'metric': 'euclidean',
    'color_palette': 'seismic',
    'margin_palette': 'cubehelix',
    'x_labels': False,
    'y_labels': False,
    'level': 3
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
matplotlib, numpy, pandas, seaborn — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ClusterGrid — a ClusterGrid output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ClusterGrid 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
read_metadata Ran
ran on a synthesized input
biocore/songbird/songbird/util.py
code served (permissive licence) · get_code("34cfd615f5f453a1")
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 numpy as np
import pandas as pd
import tempfile
import os

# Create a temporary metadata file
temp_dir = tempfile.mkdtemp()
filepath = os.path.join(temp_dir, "metadata.tsv")

# Create sample metadata content
metadata_content = """SampleID\tValue1\tValue2\tCategory
Sample1\t1.5\t2.3\tA
Sample2\t3.2\t4.1\tB
Sample3\t5.0\t6.7\tC
"""

with open(filepath, "w") as f:
    f.write(metadata_content)

ARGS = {
    "filepath": filepath
}
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
split_tables Ran
ran on a synthesized input
biocore/mmvec/mmvec/util.py
code served (permissive licence) · get_code("bd56eda3c2ff5c22")
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
from unittest.mock import Mock

# Create a mock biom.Table class
class MockBiomTable:
    def __init__(self, data_dict):
        self.data_dict = data_dict
    
    def to_dataframe(self):
        return pd.DataFrame(self.data_dict)

# Create sample data for OTU table (microbes x samples)
# We need samples as index and microbes as columns
otu_data = {
    'microbe_1': [5.0, 3.0, 0.0, 2.0, 1.0, 4.0, 0.0, 0.0, 2.0, 0.0, 1.0, 2.0],
    'microbe_2': [0.0, 0.0, 8.0, 0.0, 5.0, 0.0, 3.0, 2.0, 0.0, 4.0, 0.0, 1.0],
    'microbe_3': [2.0, 1.0, 0.0, 3.0, 0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0],
    'microbe_4': [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0],
}

otu_table_df = pd.DataFrame(otu_data, index=[f'sample_{i}' for i in range(12)])
otu_table = MockBiomTable(otu_table_df.T.to_dict('list'))

# Create sample data for metabolite table
metabolite_data = {
    'metabolite_1': [100.0, 150.0, 200.0, 120.0, 180.0, 110.0, 190.0, 130.0, 140.0, 160.0, 170.0, 125.0],
    'metabolite_2': [50.0, 60.0, 75.0, 55.0, 70.0, 52.0, 80.0, 58.0, 62.0, 72.0, 68.0, 54.0],
}

metabolite_table_df = pd.DataFrame(metabolite_data, index=[f'sample_{i}' for i in range(12)])
metabolite_table = MockBiomTable(metabolite_table_df.T.to_dict('list'))

# Create metadata DataFrame with training column
metadata = pd.DataFrame({
    'training_column': ['Train', 'Train', 'Train', 'Train', 'Train', 'Train', 
                       'Test', 'Test', 'Test', 'Test', 'Test', 'Test']
}, index=[f'sample_{i}' for i in range(12)])

ARGS = {
    'otu_table': otu_table,
    'metabolite_table': metabolite_table,
    'metadata': metadata,
    'training_column': 'training_column',
    'num_test': 3,
    'min_samples': 2
}
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
match_and_filter Not yet run
unverified (harvested, no recorded run)
biocore/songbird/songbird/util.py
code served (permissive licence) · get_code("9235ab9e7549c957")
How it ran

no invocation recorded.

match_and_filter Not yet run
unverified (harvested, no recorded run)
mortonjt/songbird/songbird/util.py
code served (permissive licence) · get_code("835a060fe4905c7b")
How it ran

no invocation recorded.

Also present in other papers' code

Of 7 samples lifted from this paper's linked repositories, 7 carry a function identity and 0 do not. Of the 7 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 7 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

Gradient descent optimization algorithms, while increasingly popular, are often used as black-box optimizers, as practical explanations of their strengths and weaknesses are hard to come by. This article aims to provide the reader with intuitions with regard to the behaviour of different algorithms that will allow her to put them to use. In the course of this overview, we look at different variants of gradient descent, summarize challenges, introduce the most common optimization algorithms, review architectures in a parallel and distributed setting, and investigate additional strategies for optimizing gradient descent.

For agents

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

get_harvested_code_for_paper("1609.04747")
get_code_for_paper("1609.04747")
have("1609.04747")

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

curl https://syntology.ai/api/ran/1609.04747.json

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

[![Syntology run record](https://syntology.ai/api/ran/1609.04747.svg)](https://syntology.ai/paper/1609.04747)

Connect an agent — have() is free.