SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2203.09137 · CVPR · 2022

Global Convergence of MAML and Theory-Inspired Neural Architecture Search for Few-Shot Learning

Bo Li, Ruoyu Sun, Haoxiang Wang, Yite Wang

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; 0 honoured the contract we drafted, 1 violated it and 0 have no contract verdict; 3 failures 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/2203.09137.json · badge: /api/ran/2203.09137.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
YiteWang/MetaNTK-NAS community
Papers with Code archive
4 ran · 0 no recorded run
FunctionStatusWhere it lives
check_symmetric Ran
ran, violated the contract we drafted
YiteWang/MetaNTK-NAS/lib/procedures/ntk_opacus.py
code served (permissive licence) · get_code("99af528f6bc21a28")
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) and returned bool, 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 numpy as np

# Create a small symmetric matrix for testing
symmetric_matrix = np.array([
    [1.0, 2.0, 3.0],
    [2.0, 4.0, 5.0],
    [3.0, 5.0, 6.0]
])

ARGS = {
    "a": symmetric_matrix,
    "rtol": 1e-05,
    "atol": 1e-08
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type bool, 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
compute_analytical_metantk_n Ran
ran; our contract draft was wrong, not the code
YiteWang/MetaNTK-NAS/lib/procedures/ntk_opacus.py
code served (permissive licence) · get_code("b1f294d5269402b1")
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) and returned dict. 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 torch
import numpy as np
import scipy

torch.manual_seed(0)

# Create small symmetric PSD matrices for ntk and K
# n_tasks=2, n_queries=2, n_support=2
# Total size: 2*(2+2) = 8

n_tasks = 2
n_queries = 2
n_support = 2
total_size = n_tasks * (n_queries + n_support)  # 8

# Create a random matrix and symmetrize it to ensure symmetry
random_matrix = np.random.randn(total_size, total_size)
ntk_np = (random_matrix + random_matrix.T) / 2

# Make it PSD by adding a large diagonal term
ntk_np = ntk_np + np.eye(total_size) * (np.abs(np.min(np.linalg.eigvalsh(ntk_np))) + 1)

# Create K matrix (for ANIL, K should be different from ntk)
random_matrix_k = np.random.randn(total_size, total_size)
K_np = (random_matrix_k + random_matrix_k.T) / 2
K_np = K_np + np.eye(total_size) * (np.abs(np.min(np.linalg.eigvalsh(K_np))) + 1)

# Convert to torch tensors
ntk = torch.from_numpy(ntk_np).float()
K = torch.from_numpy(K_np).float()

# Parameters
inner_lr_time = 0.1
reg_coef = 1e-5
algorithm = 'ANIL'

ARGS = {
    'ntk': ntk,
    'K': K,
    'n_tasks': n_tasks,
    'n_queries': n_queries,
    'n_support': n_support,
    'inner_lr_time': inner_lr_time,
    'reg_coef': reg_coef,
    'algorithm': algorithm
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy, scipy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
kernel_T_func Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
YiteWang/MetaNTK-NAS/lib/procedures/ntk_opacus.py
code served (permissive licence) · get_code("a62a0f329154799b")
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) and returned ndarray, ndim 2, dtype family float, shape (4, 4) 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 numpy as np
import scipy

# Create a small symmetric positive semi-definite matrix
np.random.seed(0)
matrix_size = 4
# Generate a random matrix and make it symmetric PSD
random_matrix = np.random.randn(matrix_size, matrix_size)
kernel = random_matrix @ random_matrix.T  # This ensures it's symmetric PSD

# Define time_evolution parameter
time_evolution = 0.5

# Define regularization coefficient
reg_coef = 1e-5

ARGS = {
    "kernel": kernel,
    "time_evolution": time_evolution,
    "reg_coef": reg_coef
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy, scipy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 2, dtype family float, shape (4, 4) — shape (4, 4) 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
split_kernel_into_blocks Ran
ran; our contract draft was wrong, not the code
YiteWang/MetaNTK-NAS/lib/procedures/ntk_opacus.py
code served (permissive licence) · get_code("16e3b53b85631e5e")
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) and returned list, ndim 4, 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 numpy as np

# Create a kernel matrix that can be evenly split
# For n_tasks=2, we need a square matrix divisible by 2 in both dimensions
# Let's create a 4x4 kernel that can be split into 2x2 blocks
kernel = np.arange(16).reshape(4, 4).astype(np.float32)

n_tasks = 2

ARGS = {
    "kernel": kernel,
    "n_tasks": n_tasks,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 4, 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, 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 4 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

Model-agnostic meta-learning (MAML) and its variants have become popular approaches for few-shot learning. However, due to the non-convexity of deep neural nets (DNNs) and the bi-level formulation of MAML, the theoretical properties of MAML with DNNs remain largely unknown. In this paper, we first prove that MAML with overparameterized DNNs is guaranteed to converge to global optima at a linear rate. Our convergence analysis indicates that MAML with over-parameterized DNNs is equivalent to kernel regression with a novel class of kernels, which we name as Meta Neural Tangent Kernels (MetaNTK). Then, we propose MetaNTK-NAS, a new training-free neural architecture search (NAS) method for few-shot learning that uses MetaNTK to rank and select architectures. Empirically, we compare our MetaNTK-NAS with previous NAS methods on two popular few-shot learning benchmarks, miniImageNet, and tieredImageNet. We show that the performance of MetaNTK-NAS is comparable or better than the state-of-the-art NAS method designed for fewshot learning while enjoying more than 100x speedup. We believe the efficiency of MetaNTK-NAS makes itself more practical for many real-world tasks. Our code is released at github.com/YiteWang/MetaNTK-NAS.

For agents

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

get_harvested_code_for_paper("2203.09137")
get_code_for_paper("2203.09137")
have("2203.09137")

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

curl https://syntology.ai/api/ran/2203.09137.json

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

[![Syntology run record](https://syntology.ai/api/ran/2203.09137.svg)](https://syntology.ai/paper/2203.09137)

Connect an agent — have() is free.