SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2402.03496 · ICML · 2024

Can We Remove the Square-Root in Adaptive Gradient Methods? A Second-Order Perspective

Richard Turner, Juhan Bae, Alireza Makhzani, Felix Dangel, Wu Lin, Runa Eschenhagen, Sgd Adamw, Rf-Rmsprop, Swinvit -Imagewoof10, Focalnet -Imagewoof10, Rmsprop Sgd Rf-Rmsprop

arXiv · PDF · Open in the Atlas

Code that ran

9 of 10 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; 3 honoured the contract we drafted, 1 violated it and 0 have no contract verdict; 5 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/2402.03496.json · badge: /api/ran/2402.03496.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
yorkerlin/remove-the-square-root canonical (classifier label)
classifier basis: census
8 ran · 1 no recorded run
f-dangel/sirfshampoo canonical (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
FunctionStatusWhere it lives
are_states_equal Ran
ran, violated the contract we drafted
yorkerlin/remove-the-square-root/distributed_shampoo/utils/optimizer_modules.py
pointer only (licence: NONE) · get_code("fbc54c48ccc1b867")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, yorkerlin/remove-the-square-root 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.

check_diagonal Ran
ran; our contract draft was wrong, not the code
yorkerlin/remove-the-square-root/distributed_shampoo/utils/matrix_functions.py
pointer only (licence: NONE) · get_code("8817b0124e846311")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, yorkerlin/remove-the-square-root 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.

count_parameters Ran
ran, honoured the contract we drafted
yorkerlin/remove-the-square-root/models/CNNs/main_search.py
pointer only (licence: NONE) · get_code("f6b944f50d3f15ae")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, yorkerlin/remove-the-square-root 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.

distribute_buffer_sizes Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
yorkerlin/remove-the-square-root/distributed_shampoo/utils/shampoo_dist_utils.py
pointer only (licence: NONE) · get_code("f3734c77315d1566")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, yorkerlin/remove-the-square-root 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_state_dict Ran
ran; our contract draft was wrong, not the code
yorkerlin/remove-the-square-root/distributed_shampoo/utils/shampoo_checkpoint_utils.py
pointer only (licence: NONE) · get_code("cc88d01d536d1d1f")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, yorkerlin/remove-the-square-root 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_batch_size Ran
ran, honoured the contract we drafted
f-dangel/sirfshampoo/sirfshampoo/optimizer.py
code served (permissive licence) · get_code("75657f6957e455f7")
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 int, ndim 0, dtype family int, 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
from typing import Tuple

torch.manual_seed(0)

# Create a tuple of input tensors
# The function accesses inputs[0].shape[0], so we need at least one tensor
# Make tensors with distinct dimensions for probing
tensor1 = torch.randn(2, 4)  # batch_size=2, feature_dim=4
tensor2 = torch.randn(2, 3)  # batch_size=2, other_dim=3

inputs = (tensor1, tensor2)

ARGS = {
    "inputs": inputs
}
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 int, ndim 0, dtype family int, 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
get_dtype_size Ran
ran, honoured the contract we drafted
yorkerlin/remove-the-square-root/distributed_shampoo/utils/shampoo_dist_utils.py
pointer only (licence: NONE) · get_code("42aea64d67efd3c6")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, yorkerlin/remove-the-square-root 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.

matrix_root_diagonal Ran
ran; our contract draft was wrong, not the code
yorkerlin/remove-the-square-root/distributed_shampoo/utils/matrix_functions.py
pointer only (licence: NONE) · get_code("c1080dc0088b4e38")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, yorkerlin/remove-the-square-root 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.

split_local_dist_buffers Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
yorkerlin/remove-the-square-root/distributed_shampoo/utils/shampoo_dist_utils.py
pointer only (licence: NONE) · get_code("63647b62fc2b7f30")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, yorkerlin/remove-the-square-root 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.

matrix_inverse_root Not yet run
unverified (harvested, no recorded run)
yorkerlin/remove-the-square-root/distributed_shampoo/utils/matrix_functions.py
pointer only (licence: NONE) · get_code("c801b0432c4ae761")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, yorkerlin/remove-the-square-root 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 10 samples lifted from this paper's linked repositories, 10 carry a function identity and 0 do not. Of the 10 distinct functions with an identity, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 1 the same month, 1 later) and 9 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
count_parameters 180 other papers
115 dated earlier, 3 the same month, 56 later, 6 same year, order not determinable · 8 only as an assembly brick
1705.08415 (2017-05)
Supervised Community Detection with Line Graph Neural Networks
1707.01926 (2017-07)
Diffusion Convolutional Recurrent Neural Network: Data-Driven Traffic Forecasting
1808.09367 (2018-08)
Deriving Machine Attention from Human Rationales
and 177 more not named here

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

Adaptive gradient optimizers like Adam(W) are the default training algorithms for many deep learning architectures, such as transformers. Their diagonal preconditioner is based on the gradient outer product which is incorporated into the parameter update via a square root. While these methods are often motivated as approximate second-order methods, the square root represents a fundamental difference. In this work, we investigate how the behavior of adaptive methods changes when we remove the root, i.e. strengthen their second-order motivation. Surprisingly, we find that such square-root-free adaptive methods close the generalization gap to SGD on convolutional architectures, while maintaining their rootbased counterpart's performance on transformers. The second-order perspective also has practical benefits for developing non-diagonal methods that can incorporate arbitrary curvature approximations through the concept of preconditioner invariance. In contrast to root-based methods like Shampoo, root-free counterparts work well and fast with half-precision since they do not require numerically unstable matrix root decompositions and inversions. Overall, our findings provide new insights into the development of adaptive methods and raise important questions regarding the overlooked role of adaptivity in their success.

For agents

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

get_harvested_code_for_paper("2402.03496")
get_code_for_paper("2402.03496")
have("2402.03496")

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

curl https://syntology.ai/api/ran/2402.03496.json

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

[![Syntology run record](https://syntology.ai/api/ran/2402.03496.svg)](https://syntology.ai/paper/2402.03496)

Connect an agent — have() is free.