SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2605.22644 · 2026

Why SGD is not Brownian Motion: A New Perspective on Stochastic Dynamics

Aleksandr Beznosikov, Aleksandr Kovalenko, Andrey Veprikov, Andrew Semenov, Igor Ignashin, Anna Radovskaya, Egor Lopatin, Ole- Govich Stanislav, Potapov, Aleksandr Shestakov, Andrey Leonidov

arXiv · PDF · Open in the Atlas

Code that ran

13 of 17 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; 4 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 9 failures of our instrument; 4 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/2605.22644.json · badge: /api/ran/2605.22644.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
brain-lab-research/SGDiffusion canonical (classifier label)
classifier basis: census
13 ran · 4 no recorded run
FunctionStatusWhere it lives
alpha_estimator Ran
ran; our contract draft was wrong, not the code
brain-lab-research/SGDiffusion/src/visualise_utils.py
pointer only (licence: NONE) · get_code("950b82351b146029")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, brain-lab-research/SGDiffusion 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.

alpha_estimator2 Ran
ran; our contract draft was wrong, not the code
brain-lab-research/SGDiffusion/src/visualise_utils.py
pointer only (licence: NONE) · get_code("3063a3c8f1afe8f6")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, brain-lab-research/SGDiffusion 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.

compute_gradient_distances Ran
ran; our contract draft was wrong, not the code
brain-lab-research/SGDiffusion/src/statistic.py
pointer only (licence: NONE) · get_code("dfdedfdb8e7de832")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, brain-lab-research/SGDiffusion 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.

flatten_grads Ran
ran, honoured the contract we drafted
brain-lab-research/SGDiffusion/src/datamodelopt/core/tensor_utils.py
pointer only (licence: NONE) · get_code("111ce930928fefee")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, brain-lab-research/SGDiffusion 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.

flatten_params Ran
ran, honoured the contract we drafted
brain-lab-research/SGDiffusion/src/datamodelopt/core/tensor_utils.py
pointer only (licence: NONE) · get_code("c7d36593f7e19c5b")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, brain-lab-research/SGDiffusion 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.

format_time Ran
ran; our contract draft was wrong, not the code
brain-lab-research/SGDiffusion/src/datamodelopt/core/logging.py
pointer only (licence: NONE) · get_code("a72e1393e15cbac1")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, brain-lab-research/SGDiffusion 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_device Ran
ran; our contract draft was wrong, not the code
brain-lab-research/SGDiffusion/src/datamodelopt/core/config.py
pointer only (licence: NONE) · get_code("340a7bfff5f1f4a5")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, brain-lab-research/SGDiffusion 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_generator Ran
ran; our contract draft was wrong, not the code
brain-lab-research/SGDiffusion/src/datamodelopt/core/seed.py
pointer only (licence: NONE) · get_code("cbc497fd8c503d83")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, brain-lab-research/SGDiffusion 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_param_count Ran
ran, honoured the contract we drafted
brain-lab-research/SGDiffusion/src/datamodelopt/core/tensor_utils.py
pointer only (licence: NONE) · get_code("95ce2645335fb4b6")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, brain-lab-research/SGDiffusion 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_torch_dtype Ran
ran; our contract draft was wrong, not the code
brain-lab-research/SGDiffusion/src/datamodelopt/core/config.py
pointer only (licence: NONE) · get_code("cb3865cbb3050b3c")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, brain-lab-research/SGDiffusion 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.

init_data Ran
ran; our contract draft was wrong, not the code
brain-lab-research/SGDiffusion/src/visualise_utils.py
pointer only (licence: NONE) · get_code("e7898b70f6748622")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, brain-lab-research/SGDiffusion 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_model Ran
ran; our contract draft was wrong, not the code
brain-lab-research/SGDiffusion/src/datamodelopt/core/checkpointing.py
pointer only (licence: NONE) · get_code("645eff8e04b31c3e")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, brain-lab-research/SGDiffusion 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.

train Ran
ran, honoured the contract we drafted
brain-lab-research/SGDiffusion/src/statistic.py
pointer only (licence: NONE) · get_code("d0d3b43ce3b9e921")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, brain-lab-research/SGDiffusion 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.

CIFAR Not yet run
unverified (harvested, no recorded run)
brain-lab-research/SGDiffusion/src/utils.py
pointer only (licence: NONE) · get_code("43368a0a5b49f9e3")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, brain-lab-research/SGDiffusion 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.

MNIST Not yet run
unverified (harvested, no recorded run)
brain-lab-research/SGDiffusion/src/utils.py
pointer only (licence: NONE) · get_code("6b3e2abd8b904f36")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, brain-lab-research/SGDiffusion 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.

evaluate_sgd_update Not yet run
unverified (harvested, no recorded run)
brain-lab-research/SGDiffusion/src/statistic.py
pointer only (licence: NONE) · get_code("27c7edb2a414117f")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, brain-lab-research/SGDiffusion 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.

load_similar_mnist_data Not yet run
unverified (harvested, no recorded run)
brain-lab-research/SGDiffusion/src/utils.py
pointer only (licence: NONE) · get_code("40f7bef53b52ad30")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, brain-lab-research/SGDiffusion 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 17 samples lifted from this paper's linked repositories, 17 carry a function identity and 0 do not. Of the 17 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 17 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

Stochastic Gradient Descent (SGD) is commonly modeled as a Langevin process, assuming that minibatch noise acts as Brownian motion. However, this approximation relies on a continuous-time limit and a √ η noise scaling that does not match the discrete SGD update at finite learning rate. In this work, we propose an alternative formulation of SGD as deterministic dynamics in a fluctuating loss landscape induced by minibatch sampling. Starting directly from the discrete update, we derive a master equation for the parameter distribution and obtain a discrete Fokker-Planck equation that differs from the standard Langevin form at order η 2 . Using this framework, we analyze SGD dynamics near critical points of the loss. We show that the behavior decomposes along the eigenbasis of the mean Hessian into qualitatively distinct regimes. In particular, nearly-flat directions do not admit a stationary distribution: the variance grows over time, corresponding to effective diffusion along valleys with a coefficient proportional to the learning rate. We provide empirical evidence supporting these predictions on neural network models in computer vision and natural language processing, observing a clear qualitative separation between confined and diffusive modes.

For agents

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

get_harvested_code_for_paper("2605.22644")
get_code_for_paper("2605.22644")
have("2605.22644")

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

curl https://syntology.ai/api/ran/2605.22644.json

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

[![Syntology run record](https://syntology.ai/api/ran/2605.22644.svg)](https://syntology.ai/paper/2605.22644)

Connect an agent — have() is free.