SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2406.01589 · ICML · 2024

Tilting the Odds at the Lottery: the Interplay of Overparameterisation and Curricula in Neural Networks

Stefano Sarao Mannelli, Luca Saglietti, Andrew Saxe, Yaraslau Ivashynka, Stefano Mannelli

arXiv · PDF · Open in the Atlas

Code that ran

2 of 3 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 2 have no contract verdict; 0 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/2406.01589.json · badge: /api/ran/2406.01589.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
qglht/repal community (classifier label)
classifier basis not recorded on the link
2 ran · 1 no recorded run
FunctionStatusWhere it lives
find_accuracy_model Ran
ran on a synthesized input
qglht/repal/src/dissimilarity_pairwise.py
code served (permissive licence) · get_code("622d0435fd04e947")
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 float, 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 os
import torch
import tempfile
import shutil

torch.manual_seed(0)

# Create a temporary directory structure with checkpoint files
temp_dir = tempfile.mkdtemp()

# Create a mock checkpoint file
checkpoint_data = {
    'log': {
        'perf_min': [0.95, 0.93, 0.92, 0.91]
    }
}

# Create 50+ checkpoint files to satisfy the condition
for i in range(50):
    checkpoint_path = os.path.join(temp_dir, f'epoch_{i}_checkpoint.pth')
    torch.save(checkpoint_data, checkpoint_path)

# Create one more checkpoint with different perf_min value
checkpoint_data_final = {
    'log': {
        'perf_min': [0.95, 0.93, 0.92, 0.91, 0.85]
    }
}
final_checkpoint_path = os.path.join(temp_dir, f'epoch_50_checkpoint.pth')
torch.save(checkpoint_data_final, final_checkpoint_path)

ARGS = {
    'name': temp_dir,
    'device': 'cpu'
}
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 float, 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
find_accuracy_model Ran
ran on a synthesized input
qglht/repal/src/dissimilarity_pairwise_mamba.py
code served (permissive licence) · get_code("14007f6f3a333c28")
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 float, 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 os
import torch
import tempfile
import shutil

torch.manual_seed(0)

# Create a temporary directory structure with checkpoint files
temp_dir = tempfile.mkdtemp()

# Create mock checkpoint files
for epoch in range(60):
    checkpoint_path = os.path.join(temp_dir, f'epoch_{epoch}_checkpoint.pth')
    checkpoint_data = {
        'log': {
            'perf_min': [0.85 + epoch * 0.001]
        }
    }
    torch.save(checkpoint_data, checkpoint_path)

device = 'cpu'
name = temp_dir

ARGS = {
    'name': name,
    'device': device
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
matplotlib, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float, 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
load_model Not yet run
unverified (harvested, no recorded run)
qglht/repal/main/model.py
code served (permissive licence) · get_code("c71208dc95adea72")
How it ran

no invocation recorded.

Also present in other papers' code

Of 3 samples lifted from this paper's linked repositories, 3 carry a function identity and 0 do not. Of the 2 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 2 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

A wide range of empirical and theoretical works have shown that overparameterisation can amplify the performance of neural networks. According to the lottery ticket hypothesis, overparameterised networks have an increased chance of containing a sub-network that is well-initialised to solve the task at hand. A more parsimonious approach, inspired by animal learning, consists in guiding the learner towards solving the task by curating the order of the examples, i.e. providing a curriculum. However, this learning strategy seems to be hardly beneficial in deep learning applications. In this work, we undertake an analytical study that connects curriculum learning and overparameterisation. In particular, we investigate their interplay in the online learning setting for a 2-layer network in the XOR-like Gaussian Mixture problem. Our results show that a high degree of overparameterisation-while simplifying the problem-can limit the benefit from curricula, providing a theoretical account of the ineffectiveness of curricula in deep learning.

For agents

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

get_harvested_code_for_paper("2406.01589")
get_code_for_paper("2406.01589")
have("2406.01589")

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

curl https://syntology.ai/api/ran/2406.01589.json

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

[![Syntology run record](https://syntology.ai/api/ran/2406.01589.svg)](https://syntology.ai/paper/2406.01589)

Connect an agent — have() is free.