SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2106.06695 · ICML · 2021

SKIing on Simplices: Kernel Interpolation on the Permutohedral Lattice for Scalable Gaussian Processes

Marc Finzi, Andrew Wilson, Sanyam Kapoor, Ke Wang

arXiv · PDF · Open in the Atlas

Code that ran

11 of 13 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 11 have no contract verdict; 0 failures of our instrument; 2 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/2106.06695.json · badge: /api/ran/2106.06695.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
activatedgeek/simplex-gp canonical (classifier label)
classifier basis: census
11 ran · 2 no recorded run
FunctionStatusWhere it lives
binary_search Ran
ran on a synthesized input
activatedgeek/simplex-gp/gpytorch_lattice_kernel/bilateral_kernel.py
code served (permissive licence) · get_code("531d10e023e4f5ad")
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 torch
import numpy as np

torch.manual_seed(0)

# Define a simple monotonic function for binary search
# This function will be used as the 'fn' parameter
def monotonic_fn(x):
    """A simple monotonic function: y = 2*x + 1"""
    return 2 * x + 1

# Set up the parameters for binary_search
target = 5.0  # We want to find x such that fn(x) ≈ 5.0
# For fn(x) = 2*x + 1, when target = 5.0, x should be around 2.0

bounds = (0.0, 10.0)  # Initial search bounds [lower, upper]

fn = monotonic_fn  # The monotonic function to search with

eps = 0.01  # Precision threshold

ARGS = {
    "target": target,
    "bounds": bounds,
    "fn": fn,
    "eps": eps
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, 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
compute Ran
ran on a synthesized input
activatedgeek/simplex-gp/experiments/mvm_err.py
code served (permissive licence) · get_code("621e361c396ba3b5")
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 tuple. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is 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 torch.nn as nn

torch.manual_seed(0)

class KernelModule(nn.Module):
    """Callable kernel that takes two tensors and returns a matrix."""
    def forward(self, X1, X2):
        # Simple kernel: returns a square matrix of shape (X1.shape[0], X2.shape[0])
        # This mimics a kernel matrix computation
        return torch.randn(X1.shape[0], X2.shape[0])

K = KernelModule()
X = torch.randn(2, 4)  # batch_size=2, feature_dim=4
y = torch.randn(2, 3)  # batch_size=2, output_dim=3

ARGS = {
    "K": K,
    "X": X,
    "y": y,
    "n_iter": 5
}
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 tuple — a tuple output has no array shape of its own; none is 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
cos_err Ran
ran on a synthesized input
activatedgeek/simplex-gp/experiments/mvm_err.py
code served (permissive licence) · get_code("5119636d565182a5")
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

torch.manual_seed(0)

# cos_err computes cosine similarity between two vectors x and y
# It requires:
# - x: a tensor with .sum() and .norm(p=2) methods
# - y: a tensor with .sum() and .norm(p=2) methods
# Both should be 1D or multi-dimensional tensors that support element-wise multiplication

x = torch.randn(4)
y = torch.randn(4)

ARGS = {
    "x": x,
    "y": y,
}
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
coverage_diff Ran
ran on a synthesized input
activatedgeek/simplex-gp/gpytorch_lattice_kernel/bilateral_kernel.py
code served (permissive licence) · get_code("72b330493715d4a3")
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 float64, ndim 0, 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
import logging

logging.basicConfig(level=logging.INFO)

# Analyze the function to understand what each parameter needs to be:
# - spacing: float, used in arithmetic operations
# - order: int, used in arithmetic (2 * order + 1)
# - x: array-like, used with numpy comparison operators and indexing
# - w: array-like, used with numpy comparison operators and indexing
# - fn_values: array-like, indexed by boolean mask and summed
# - fft_values: array-like, indexed by boolean mask and summed

# Create small numpy arrays for the function parameters
spacing = 0.5  # small float value
order = 2  # small integer

# Create x array with values in a range that will allow meaningful filtering
x = np.array([-3.0, -1.5, 0.0, 1.5, 3.0, 4.5])

# Create w array (frequency domain)
w = np.array([-8.0, -4.0, 0.0, 4.0, 8.0, 10.0])

# Create fn_values (spatial domain function values)
# Should have same length as x for indexing to work
fn_values = np.array([0.1, 0.3, 0.5, 0.4, 0.2, 0.05])

# Create fft_values (frequency domain function values)
# Should have same length as w for indexing to work
fft_values = np.array([0.05, 0.2, 0.6, 0.15, 0.08, 0.02])

ARGS = {
    'spacing': spacing,
    'order': order,
    'x': x,
    'w': w,
    'fn_values': fn_values,
    'fft_values': fft_values
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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
generate_sweep Ran
ran on a synthesized input
activatedgeek/simplex-gp/experiments/wandb_utils.py
code served (permissive licence) · get_code("d0d692923320cfee")
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 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 torch.nn as nn
from datetime import datetime

torch.manual_seed(0)

ARGS = {
    'alg': 'bigp',
    'dataset': 'mnist',
    'lr': 0.05,
    'lanc_iter': 50,
    'pre_size': 10,
    'project': 'bilateral-gp-experiments',
    'submit': False
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, wandb — 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
get_coeffs Ran
ran on a synthesized input
activatedgeek/simplex-gp/gpytorch_lattice_kernel/bilateral_kernel.py
code served (permissive licence) · get_code("f712aa8240e39595")
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 1, dtype family float, shape (5,) 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 numpy as np
from functools import partial

torch.manual_seed(0)

# Create a simple kernel function that takes a tensor and returns a tensor
def simple_kernel_fn(x):
    """A simple stationary isotropic kernel function: Gaussian-like."""
    return torch.exp(-x**2 / 2)

ARGS = {
    "kernel_fn": simple_kernel_fn,
    "order": 2
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 1, dtype family float, shape (5,) — shape (5,) 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
rel_err Ran
ran on a synthesized input
activatedgeek/simplex-gp/experiments/mvm_err.py
code served (permissive licence) · get_code("01f8d06eb16ad4d5")
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

torch.manual_seed(0)

# Create two small tensors for comparison
x = torch.randn(2, 4)
y = torch.randn(2, 4)

ARGS = {
    'x': x,
    'y': y
}
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
test Ran
ran on a synthesized input
activatedgeek/simplex-gp/experiments/train_simplexgp.py
code served (permissive licence) · get_code("224de20ec1631761")
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 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 gpytorch as gp
from gpytorch.models import ExactGP
from gpytorch.means import ConstantMean
from gpytorch.kernels import ScaleKernel, RBFKernel
from gpytorch.distributions import MultivariateNormal
from gpytorch.likelihoods import GaussianLikelihood

torch.manual_seed(0)

# Create tiny test data
x = torch.randn(2, 4)  # batch_size=2, input_dim=4
y = torch.randn(2)     # batch_size=2

# Create a minimal GPyTorch model
class TinyGPModel(ExactGP):
    def __init__(self, train_x, train_y, likelihood):
        super().__init__(train_x, train_y, likelihood)
        self.mean_module = ConstantMean()
        self.covar_module = ScaleKernel(RBFKernel())
    
    def forward(self, x):
        mean_x = self.mean_module(x)
        covar_x = self.covar_module(x)
        return MultivariateNormal(mean_x, covar_x)

likelihood = GaussianLikelihood()
model = TinyGPModel(x, y, likelihood)

# Create MLL (marginal log likelihood)
mll = gp.mlls.ExactMarginalLogLikelihood(likelihood, model)

ARGS = {
    'x': x,
    'y': y,
    'model': model,
    'mll': mll,
    'lanc_iter': 100,
    'pre_size': 100,
    'cg_iter': 1000,
    'cg_tol': 0.01,
    'label': 'test'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
gpytorch, 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
train Ran
ran on a synthesized input
activatedgeek/simplex-gp/experiments/train_keops.py
code served (permissive licence) · get_code("6314d25229de28ca")
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 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 gpytorch as gp
from gpytorch.models import ExactGP
from gpytorch.means import ConstantMean
from gpytorch.kernels import ScaleKernel, RBFKernel
from gpytorch.distributions import MultivariateNormal
from gpytorch.likelihoods import GaussianLikelihood

torch.manual_seed(0)

# Create tiny training data
x = torch.randn(2, 4)  # batch_size=2, input_dim=4
y = torch.randn(2)      # batch_size=2, output_dim=1

# Create a minimal GPyTorch model
class MinimalGP(ExactGP):
    def __init__(self, train_x, train_y, likelihood):
        super().__init__(train_x, train_y, likelihood)
        self.mean_module = ConstantMean()
        self.covar_module = ScaleKernel(RBFKernel())
    
    def forward(self, x):
        mean_x = self.mean_module(x)
        covar_x = self.covar_module(x)
        return MultivariateNormal(mean_x, covar_x)

likelihood = GaussianLikelihood()
model = MinimalGP(x, y, likelihood)

# Create marginal log likelihood
mll = gp.mlls.ExactMarginalLogLikelihood(likelihood, model)

# Create optimizer
optim = torch.optim.Adam(model.parameters(), lr=0.01)

ARGS = {
    'x': x,
    'y': y,
    'model': model,
    'mll': mll,
    'optim': optim,
    'lanc_iter': 100,
    'pre_size': 100
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
gpytorch, 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
train Ran
ran on a synthesized input
activatedgeek/simplex-gp/experiments/train_simplexgp.py
code served (permissive licence) · get_code("a7ef504140137c14")
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 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 gpytorch as gp
from gpytorch.models import ExactGP
from gpytorch.means import ConstantMean
from gpytorch.kernels import ScaleKernel, RBFKernel
from gpytorch.distributions import MultivariateNormal
from gpytorch.likelihoods import GaussianLikelihood
import torch.optim as optim

torch.manual_seed(0)

# Create small tensors
x = torch.randn(2, 4)  # batch=2, dim=4
y = torch.randn(2)     # batch=2

# Create a minimal GPyTorch model
class SimpleGPModel(ExactGP):
    def __init__(self, train_x, train_y, likelihood):
        super(SimpleGPModel, self).__init__(train_x, train_y, likelihood)
        self.mean_module = ConstantMean()
        self.covar_module = ScaleKernel(RBFKernel())

    def forward(self, x):
        mean_x = self.mean_module(x)
        covar_x = self.covar_module(x)
        return MultivariateNormal(mean_x, covar_x)

# Create likelihood and model
likelihood = GaussianLikelihood()
model = SimpleGPModel(x, y, likelihood)

# Create MLL (Marginal Log Likelihood)
mll = gp.mlls.ExactMarginalLogLikelihood(likelihood, model)

# Create optimizer
optimizer = optim.Adam(model.parameters(), lr=0.1)

ARGS = {
    'x': x,
    'y': y,
    'model': model,
    'mll': mll,
    'optim': optimizer,
    'lanc_iter': 100,
    'cg_iter': 1000,
    'pre_size': 100,
    'cg_tol': 1.0
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
gpytorch, 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
train Ran
ran on a synthesized input
activatedgeek/simplex-gp/experiments/train_skip.py
code served (permissive licence) · get_code("4d9c7f512a00d0e1")
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 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 gpytorch as gp
from gpytorch.models import ExactGP
from gpytorch.means import ConstantMean
from gpytorch.kernels import ScaleKernel, RBFKernel
from gpytorch.distributions import MultivariateNormal
from gpytorch.likelihoods import GaussianLikelihood
from torch import optim as torch_optim

torch.manual_seed(0)

# Create tiny tensors: batch_size=2, input_dim=4
x = torch.randn(2, 4)
y = torch.randn(2)

# Build a minimal GPyTorch model
class TinyGP(ExactGP):
    def __init__(self, train_x, train_y, likelihood):
        super().__init__(train_x, train_y, likelihood)
        self.mean_module = ConstantMean()
        self.covar_module = ScaleKernel(RBFKernel())
    
    def forward(self, x):
        mean_x = self.mean_module(x)
        covar_x = self.covar_module(x)
        return MultivariateNormal(mean_x, covar_x)

likelihood = GaussianLikelihood()
model = TinyGP(x, y, likelihood)

# Create marginal log likelihood
mll = gp.mlls.ExactMarginalLogLikelihood(likelihood, model)

# Create optimizer
optim = torch_optim.Adam(model.parameters(), lr=0.01)

ARGS = {
    'x': x,
    'y': y,
    'model': model,
    'mll': mll,
    'optim': optim,
    'lanc_iter': 100,
    'pre_size': 100
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
gpytorch, 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
test Not yet run
unverified (harvested, no recorded run)
activatedgeek/simplex-gp/experiments/train_keops.py
code served (permissive licence) · get_code("2644758dc97dafb9")
How it ran

no invocation recorded.

test Not yet run
unverified (harvested, no recorded run)
activatedgeek/simplex-gp/experiments/train_skip.py
code served (permissive licence) · get_code("c8315a839146cf9f")
How it ran

no invocation recorded.

Also present in other papers' code

Of 13 samples lifted from this paper's linked repositories, 13 carry a function identity and 0 do not. Of the 13 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 13 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

State-of-the-art methods for scalable Gaussian processes use iterative algorithms, requiring fast matrix vector multiplies (MVMs) with the covariance kernel. The Structured Kernel Interpolation (SKI) framework accelerates these MVMs by performing efficient MVMs on a grid and interpolating back to the original space. In this work, we develop a connection between SKI and the permutohedral lattice used for highdimensional fast bilateral filtering. Using a sparse simplicial grid instead of a dense rectangular one, we can perform GP inference exponentially faster in the dimension than SKI. Our approach, Simplex-GP, enables scaling SKI to high dimensions, while maintaining strong predictive performance. We additionally provide a CUDA implementation of Simplex-GP, which enables significant GPU acceleration of MVM based inference.

For agents

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

get_harvested_code_for_paper("2106.06695")
get_code_for_paper("2106.06695")
have("2106.06695")

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

curl https://syntology.ai/api/ran/2106.06695.json

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

[![Syntology run record](https://syntology.ai/api/ran/2106.06695.svg)](https://syntology.ai/paper/2106.06695)

Connect an agent — have() is free.