SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1911.02590 · 2019

Optimizing Millions of Hyperparameters by Implicit Differentiation

arXiv · PDF · Open in the Atlas

Code that ran

4 of 4 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; 1 honoured the contract we drafted, 1 violated it and 1 has no contract verdict; 1 failure of our instrument; 0 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/1911.02590.json · badge: /api/ran/1911.02590.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
ucr-optml/fedmsa community (classifier label)
classifier basis not recorded on the link
3 ran · 0 no recorded run
ucr-optml/FedNest extension (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
FunctionStatusWhere it lives
gather_flat_grad Ran
ran, honoured the contract we drafted
ucr-optml/fedmsa/core/function.py
pointer only (licence: MIT); the stored copy get_code serves, from Tarzanagh/FedNest, is pointer only (licence: NONE) · get_code("739afdca8b2d673a")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from Tarzanagh/FedNest, and GitHub detected no licence in that repository 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.

loss_adjust_cross_entropy Ran
ran on a synthesized input
ucr-optml/fedmsa/core/function.py
code served (permissive licence) · get_code("2b5f9d39a738d9b9")
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 Tensor, ndim 0, dtype family float, 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
import torch.nn as nn
import torch.nn.functional as F

torch.manual_seed(0)

# Construct arguments for loss_adjust_cross_entropy
batch_size = 2
num_classes = 4

# logits: model output with shape (batch_size, num_classes)
logits = torch.randn(batch_size, num_classes)

# targets: class indices with shape (batch_size,)
targets = torch.tensor([0, 2], dtype=torch.long)

# params: dict with 'dy' and 'ly', optionally with weight 'wy'
# dy and ly should have shape (num_classes,) for element-wise operations
dy = torch.randn(num_classes)
ly = torch.randn(num_classes)

# Optional: weight for cross_entropy with shape (num_classes,)
wy = torch.ones(num_classes)

params = {
    'dy': dy,
    'ly': ly,
    2: wy  # Using index 2 as the third parameter key
}

# group_size parameter
group_size = 1

ARGS = {
    'logits': logits,
    'targets': targets,
    'params': params,
    'group_size': group_size
}
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 Tensor, ndim 0, dtype family float, 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
neumann_hyperstep_preconditioner Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
ucr-optml/fedmsa/core/function.py
pointer only (licence: MIT); the stored copy get_code serves, from Tarzanagh/FedNest, is pointer only (licence: NONE) · get_code("085767456c708d9d")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from Tarzanagh/FedNest, and GitHub detected no licence in that repository 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.

smooth Ran
ran, violated the contract we drafted
ucr-optml/FedNest/reproduce/fig2.py
code served (permissive licence) · get_code("97aa6390755cc0a7")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Sonnet 4.5, prompt python_fixture_v1) and returned list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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

torch.manual_seed(0)

scalars = [1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0]
weight = 0.6

ARGS = {
    'scalars': scalars,
    'weight': weight
}
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
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

Also present in other papers' code

Of 4 samples lifted from this paper's linked repositories, 4 carry a function identity and 0 do not. Of the 4 distinct functions with an identity, 4 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 4 later) and 0 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
gather_flat_grad 2 other papers
2 later
2201.01212 (2022-01)
AutoBalance: Optimized Loss Functions for Imbalanced Data
2205.02215 (2022-05)
FEDNEST: Federated Bilevel, Minimax, and Compositional Optimization
neumann_hyperstep_preconditioner 2 other papers
2 later
2201.01212 (2022-01)
AutoBalance: Optimized Loss Functions for Imbalanced Data
2205.02215 (2022-05)
FEDNEST: Federated Bilevel, Minimax, and Compositional Optimization
smooth 2 other papers
2 later
2205.02215 (2022-05)
FEDNEST: Federated Bilevel, Minimax, and Compositional Optimization
2406.00392 (2024-06)
Artificial Generational Intelligence: Cultural Accumulation in Reinforcement Learning
loss_adjust_cross_entropy 1 other paper
1 later
2205.02215 (2022-05)
FEDNEST: Federated Bilevel, Minimax, and Compositional Optimization

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

We propose an algorithm for inexpensive gradient-based hyperparameter optimization that combines the implicit function theorem (IFT) with efficient inverse Hessian approximations. We present results about the relationship between the IFT and differentiating through optimization, motivating our algorithm. We use the proposed approach to train modern network architectures with millions of weights and millions of hyper-parameters. For example, we learn a data-augmentation network - where every weight is a hyperparameter tuned for validation performance - outputting augmented training examples. Jointly tuning weights and hyperparameters with our approach is only a few times more costly in memory and compute than standard training.

For agents

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

get_harvested_code_for_paper("1911.02590")
get_code_for_paper("1911.02590")
have("1911.02590")

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

curl https://syntology.ai/api/ran/1911.02590.json

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

[![Syntology run record](https://syntology.ai/api/ran/1911.02590.svg)](https://syntology.ai/paper/1911.02590)

Connect an agent — have() is free.