SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1906.01549 · 2019

Streaming Variational Monte Carlo

arXiv · PDF · Open in the Atlas

Code that ran

11 of 12 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; 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/1906.01549.json · badge: /api/ran/1906.01549.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
catniplab/svmc canonical (classifier label)
classifier basis: census
11 ran · 1 no recorded run
FunctionStatusWhere it lives
compute_elbo Ran
ran on a synthesized input
catniplab/svmc/svmc/utils.py
code served (permissive licence) · get_code("1ecb7da24d212c0f")
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 (2,) 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 a small tensor with shape (batch=2, dim=4)
# The function expects weights to support:
# - torch.mean(weights, -1) -> reduces last dimension
# - torch.cumprod(w_mean, 0) -> cumulative product along first dimension
# - indexing and arithmetic operations

weights = torch.randn(2, 4)

ARGS = {
    "weights": weights
}
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 1, dtype family float, shape (2,) — shape (2,) 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_ess Ran
ran on a synthesized input
catniplab/svmc/svmc/utils.py
code served (permissive licence) · get_code("7a4260cd0bd651fc")
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 (2,) 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 a small tensor of weights with shape (2, 4)
# The function computes ESS (effective sample size) by summing weights^2 along the last dimension
weights = torch.rand(2, 4)

ARGS = {
    "weights": weights
}
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 1, dtype family float, shape (2,) — shape (2,) 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_mse Ran
ran on a synthesized input
catniplab/svmc/svmc/utils.py
code served (permissive licence) · get_code("772924a47a39b24d")
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 (2,) 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)

# X_true: shape (batch_size, feature_dim)
# X_particles: shape (batch_size, num_particles, feature_dim)
# w_norm: shape (batch_size, num_particles) - normalized weights

batch_size = 2
num_particles = 3
feature_dim = 4

X_true = torch.randn(batch_size, feature_dim)
X_particles = torch.randn(batch_size, num_particles, feature_dim)
w_norm = torch.softmax(torch.randn(batch_size, num_particles), dim=1)

ARGS = {
    "X_true": X_true,
    "X_particles": X_particles,
    "w_norm": w_norm,
}
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 1, dtype family float, shape (2,) — shape (2,) 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
gaussian_loss Ran
ran on a synthesized input
catniplab/svmc/svmc/metric.py
code served (permissive licence) · get_code("78dd1d5f3d31dc1b")
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 (2,) 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)

# Create small tensors with distinct dimensions
batch_size = 2
dim = 4

x = torch.randn(batch_size, dim)
mu = torch.randn(batch_size, dim)
logvar = torch.randn(batch_size, dim)

ARGS = {
    "x": x,
    "mu": mu,
    "logvar": logvar,
}
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 (2,) — shape (2,) 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
multinomial_resampling Ran
ran on a synthesized input
catniplab/svmc/svmc/resample.py
code served (permissive licence) · get_code("5ca857ec7f22ea83")
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 ndarray, ndim 1, dtype family int, shape (10,) 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 numpy.random as npr

# Set seed for reproducibility
npr.seed(42)

# Create a weight array that sums to 1 (valid probability distribution)
# Using 5 particles/samples
w = np.array([0.1, 0.2, 0.3, 0.25, 0.15])

# Number of resamples to draw
M = 10

ARGS = {
    "w": w,
    "M": M
}
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 ndarray, ndim 1, dtype family int, shape (10,) — shape (10,) 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
predict Ran
ran on a synthesized input
catniplab/svmc/svmc/gp.py
code served (permissive licence) · get_code("af3fade212a88833")
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, ndim 3, 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 torch.nn as nn

torch.manual_seed(0)

# The function uses cov_func as a callable that computes covariance matrices
# It's called as: cov_func(x), cov_func(x, inducing), cov_func(inducing)
# Each returns a matrix that gets diagonal extracted or used in matrix ops

class SimpleCovFunc(nn.Module):
    def __init__(self):
        super().__init__()
    
    def forward(self, x1, x2=None):
        """Simple covariance function - returns identity-like matrices"""
        if x2 is None:
            # Kff or Kzz: (n, n) or (m, m)
            n = x1.shape[0]
            return torch.eye(n, dtype=torch.float32)
        else:
            # Kfz: (n, m)
            return torch.ones(x1.shape[0], x2.shape[0], dtype=torch.float32)

# Build yakron function (Kronecker product wrapper)
# The function calls yakron(K, A) and yakron(K, B)
def yakron(K, A):
    """Kronecker product K ⊗ A"""
    # K: (p, p), A: (n, m, m) -> result: (n*p, m*p) or similar
    p = K.shape[0]
    n, m, _ = A.shape
    result = torch.kron(K, A)
    return result

# Make yakron available in the function's scope
import sys
current_module = sys.modules[__name__]
setattr(current_module, 'yakron', yakron)

# Build arguments
# x: (n, q) - n=2 samples, q=4 dimensions
x = torch.randn(2, 4, dtype=torch.float32)

# z: (n, pm, 1) - n=2, pm=6 (product of p=3, m=2)
z = torch.randn(2, 6, 1, dtype=torch.float32)

# inducing: (m, q) - m=2 inducing points, q=4 dimensions
inducing = torch.randn(2, 4, dtype=torch.float32)

# K: (p, p) - p=3
K = torch.eye(3, dtype=torch.float32)

# cov_func: a callable that computes covariance
cov_func = SimpleCovFunc()

# gamma: (n, pm, 1) - optional, same shape as z
gamma = torch.randn(2, 6, 1, dtype=torch.float32)

ARGS = {
    'x': x,
    'z': z,
    'inducing': inducing,
    'K': K,
    'cov_func': cov_func,
    'gamma': gamma,
}
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, ndim 3, dtype family float — a tuple 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
scaled_dist Ran
ran on a synthesized input
catniplab/svmc/svmc/operation.py
code served (permissive licence) · get_code("61b64712b0ea9b2b")
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 2, dtype family float, shape (2, 3) 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 small tensors with distinct dimensions
# a: (2, 4) - batch of 2 points in 4D space
# b: (3, 4) - batch of 3 points in 4D space
# gamma: (4,) - scaling factor per dimension

a = torch.randn(2, 4)
b = torch.randn(3, 4)
gamma = torch.ones(4)

ARGS = {
    'a': a,
    'b': b,
    'gamma': gamma
}
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 2, dtype family float, shape (2, 3) — shape (2, 3) 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
sqrt Ran
ran on a synthesized input
catniplab/svmc/svmc/operation.py
code served (permissive licence) · get_code("c44d1481a7bc874c")
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 2, dtype family float, shape (2, 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 torch

torch.manual_seed(0)

# The function takes a tensor x and an optional eps parameter
# It performs a safe square root by adding eps before taking sqrt
# to avoid numerical instability with small values

x = torch.randn(2, 4)
eps = 1e-12

ARGS = {
    "x": x,
    "eps": eps
}
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 2, dtype family float, shape (2, 4) — shape (2, 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
squared_scaled_dist Ran
ran on a synthesized input
catniplab/svmc/svmc/operation.py
code served (permissive licence) · get_code("9c87125da6d62481")
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 2, dtype family float, shape (2, 3) 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 tensors with shape (batch, dim)
# Using distinct batch sizes and dims to make shapes unambiguous
a = torch.randn(2, 4)  # shape: (2, 4)
b = torch.randn(3, 4)  # shape: (3, 4) - different batch size
gamma = torch.tensor(0.5)  # scalar scaling factor

ARGS = {
    'a': a,
    'b': b,
    'gamma': gamma
}
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 2, dtype family float, shape (2, 3) — shape (2, 3) 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
stratified_resampling Ran
ran on a synthesized input
catniplab/svmc/svmc/resample.py
code served (permissive licence) · get_code("f48d34a9350409b5")
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 ndarray, ndim 1, dtype family int, 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 numpy as np
import numpy.random as npr

# Set seed for reproducibility
npr.seed(42)
np.random.seed(42)

# Create small valid arguments for stratified_resampling
w = np.array([0.1, 0.3, 0.2, 0.4])  # weights that sum to 1.0
w = w / np.sum(w)  # normalize to ensure they sum to 1.0
M = 5  # number of samples to resample

ARGS = {
    "w": w,
    "M": M
}
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 ndarray, ndim 1, dtype family int, 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
systematic_resampling Ran
ran on a synthesized input
catniplab/svmc/svmc/resample.py
code served (permissive licence) · get_code("bf1c5cf36cc90574")
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 ndarray, ndim 1, dtype family int, shape (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 numpy.random as npr

# Set seed for reproducibility
npr.seed(42)
np.random.seed(42)

# Construct arguments for systematic_resampling
# w: weights array (should sum to 1 or be normalized)
# M: number of samples to draw

w = np.array([0.2, 0.3, 0.25, 0.25])  # weights that sum to 1
M = 4  # number of resamples

ARGS = {
    "w": w,
    "M": M
}
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 ndarray, ndim 1, dtype family int, shape (4,) — shape (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
update Not yet run
unverified (harvested, no recorded run)
catniplab/svmc/svmc/gp.py
code served (permissive licence) · get_code("31e15502151f7bc6")
How it ran

no invocation recorded.

Also present in other papers' code

Of 12 samples lifted from this paper's linked repositories, 12 carry a function identity and 0 do not. Of the 12 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 12 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

Nonlinear state-space models are powerful tools to describe dynamical structures in complex time series. In a streaming setting where data are processed one sample at a time, simultaneous inference of the state and its nonlinear dynamics has posed significant challenges in practice. We develop a novel online learning framework, leveraging variational inference and sequential Monte Carlo, which enables flexible and accurate Bayesian joint filtering. Our method provides an approximation of the filtering posterior which can be made arbitrarily close to the true filtering distribution for a wide class of dynamics models and observation models. Specifically, the proposed framework can efficiently approximate a posterior over the dynamics using sparse Gaussian processes, allowing for an interpretable model of the latent dynamics. Constant time complexity per sample makes our approach amenable to online learning scenarios and suitable for real-time applications.

For agents

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

get_harvested_code_for_paper("1906.01549")
get_code_for_paper("1906.01549")
have("1906.01549")

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

curl https://syntology.ai/api/ran/1906.01549.json

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

[![Syntology run record](https://syntology.ai/api/ran/1906.01549.svg)](https://syntology.ai/paper/1906.01549)

Connect an agent — have() is free.