SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2007.05994 · ICML · 2020

State Space Expectation Propagation: Efficient Inference Schemes for Temporal Gaussian Processes

William Wilkinson, Paul Chang, Michael Andersen, Arno Solin

arXiv · PDF · Open in the Atlas

Code that ran

8 of 10 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 8 have no contract verdict; 0 failures of our instrument; 2 no recorded run), measured 2026-11-22; 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/2007.05994.json · badge: /api/ran/2007.05994.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
AaltoML/kalman-jax canonical (classifier label)
classifier basis: census
8 ran · 2 no recorded run
FunctionStatusWhere it lives
ensure_positive_precision Ran
ran on a synthesized input
AaltoML/kalman-jax/kalmanjax/utils.py
code served (permissive licence) · get_code("a2ea6b5d4192f156")
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 ArrayImpl, 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 jax.numpy as np

# The function takes a matrix K and checks its diagonal elements
# If any diagonal element is negative, it replaces negative diagonal elements with 0.01
# K should be a 2D array/matrix

# Create a small test matrix with some negative diagonal elements
K = np.array([
    [1.0, 0.2, 0.3, 0.1],
    [0.2, -0.5, 0.1, 0.0],
    [0.3, 0.1, 2.0, 0.2],
    [0.1, 0.0, 0.2, -1.5]
])

ARGS = {
    "K": K
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jax — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ArrayImpl, 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
inv Ran
ran on a synthesized input
AaltoML/kalman-jax/kalmanjax/utils.py
code served (permissive licence) · get_code("b66ab02ba6b80cfa")
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 ArrayImpl, 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 jax.numpy as jnp
import numpy as np

# Create a small positive semi-definite matrix
# We'll use a simple approach: create a matrix and multiply by its transpose
size = 4
A = np.random.randn(size, size)
P = A @ A.T  # This ensures P is PSD

# Convert to JAX array
P_jax = jnp.array(P)

ARGS = {
    "P": P_jax
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jax, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ArrayImpl, 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
log_npdf Ran
ran on a synthesized input
AaltoML/kalman-jax/kalmanjax/archive/likelihoods.py
code served (permissive licence) · get_code("18319919ff2979c6")
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 ArrayImpl, ndim 1, dtype family float, 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 jax.numpy as jnp
import numpy as np

# log_npdf computes log probability density function for a normal distribution
# It takes three parameters:
# x: the value at which to evaluate the pdf
# m: the mean of the distribution
# v: the variance of the distribution
# All three should be array-like and broadcastable

# Create small test arrays with distinct shapes for disambiguation
x = jnp.array([1.0, 2.0, 3.0, 4.0])  # shape (4,)
m = jnp.array([0.5, 1.5, 2.5, 3.5])  # shape (4,) - same as x for broadcasting
v = jnp.array([0.1, 0.2, 0.3, 0.4])  # shape (4,) - variance values

ARGS = {
    "x": x,
    "m": m,
    "v": v
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jax, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ArrayImpl, ndim 1, dtype family float, 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
mog Ran
ran on a synthesized input
AaltoML/kalman-jax/kalmanjax/archive/mog.py
code served (permissive licence) · get_code("a11de5c5e5e3520f")
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 float, shape (1000,) 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

omega = 0.8
var1 = 0.3
var2 = 0.5

def npdf(x, m, v):
    return np.exp(-(x - m) ** 2 / (2 * v)) / np.sqrt(2 * np.pi * v)

y = 0.0
f = np.linspace(-5.0, 5.0, num=1000)

def mog(y, f):
    return npdf(y, f - omega, var1) + npdf(y, f + omega, var2)

# Setup valid arguments for mog function
ARGS = {
    "y": 0.5,
    "f": np.linspace(-5.0, 5.0, num=1000)
}
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 float, shape (1000,) — shape (1000,) 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
npdf Ran
ran on a synthesized input
AaltoML/kalman-jax/kalmanjax/archive/likelihoods.py
code served (permissive licence) · get_code("886f48cb1a566d98")
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 ArrayImpl, ndim 1, dtype family float, 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 jax.numpy as jnp

# Construct arguments for npdf function
# The function computes normal probability density function values
# Parameters: x (input), m (mean), v (variance)

# Create small test arrays using jax.numpy
x = jnp.array([0.0, 1.0, 2.0, -1.0])
m = jnp.array(0.5)
v = jnp.array(1.0)

ARGS = {
    "x": x,
    "m": m,
    "v": v
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jax — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ArrayImpl, ndim 1, dtype family float, 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
npdf Ran
ran on a synthesized input
AaltoML/kalman-jax/kalmanjax/archive/mog.py
code served (permissive licence) · get_code("ef0f92280a28ff98")
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 float, 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

# The function npdf computes a normal probability density function
# It takes three parameters:
# - x: the value(s) at which to evaluate the PDF
# - m: the mean
# - v: the variance

# Create small numpy arrays for each parameter
x = np.array([0.5, 1.5, 2.5, 3.5])  # 4 values
m = np.array([1.0, 2.0, 3.0, 4.0])  # 4 means
v = np.array([0.5, 1.0, 1.5, 2.0])  # 4 variances

ARGS = {
    "x": x,
    "m": m,
    "v": v
}
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 float, 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
solve Ran
ran on a synthesized input
AaltoML/kalman-jax/kalmanjax/utils.py
code served (permissive licence) · get_code("89e6795a9c2df510")
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 ArrayImpl, ndim 2, dtype family float, shape (4, 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 jax
import jax.numpy as jnp
import numpy as np

# Set seed for reproducibility
np.random.seed(0)

# Create small PSD matrices for testing
# P should be a positive semi-definite matrix (square)
# Q can be a matrix or vector

# Create a random matrix and make it PSD by computing A @ A.T
A = np.random.randn(4, 4)
P = A @ A.T  # This is guaranteed to be PSD

# Create Q as a random matrix (4x3 for distinct dimensions)
Q = np.random.randn(4, 3)

# Convert to JAX arrays
P_jax = jnp.array(P, dtype=jnp.float32)
Q_jax = jnp.array(Q, dtype=jnp.float32)

ARGS = {
    "P": P_jax,
    "Q": Q_jax,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jax, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ArrayImpl, ndim 2, dtype family float, shape (4, 3) — shape (4, 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
threshold_func Ran
ran on a synthesized input
AaltoML/kalman-jax/kalmanjax/archive/likelihoods.py
code served (permissive licence) · get_code("917320e7cf79cb5c")
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 ArrayImpl, 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 jax.numpy as jnp
import numpy as np

torch_seed = 0

# Analyze the function: threshold_func(x, p)
# x: appears to be a numeric value/array (used in arithmetic operations)
# p: appears to be a probability/proportion value between 0 and 1 (used in multiplication)

# Small test values
x_val = jnp.array(2.5)
p_val = jnp.array(0.3)

ARGS = {
    "x": x_val,
    "p": p_val,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jax, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ArrayImpl, 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
kalman_filter Not yet run
unverified (harvested, no recorded run)
AaltoML/kalman-jax/kalmanjax/archive/kf_no_jit.py
code served (permissive licence) · get_code("f541ffe2aaec2ad5")
How it ran

no invocation recorded.

rauch_tung_striebel_smoother Not yet run
unverified (harvested, no recorded run)
AaltoML/kalman-jax/kalmanjax/archive/kf_no_jit.py
code served (permissive licence) · get_code("5a692ba4fe8d9aab")
How it ran

no invocation recorded.

Also present in other papers' code

Of 10 samples lifted from this paper's linked repositories, 10 carry a function identity and 0 do not. Of the 10 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 10 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

We formulate approximate Bayesian inference in non-conjugate temporal and spatio-temporal Gaussian process models as a simple parameter update rule applied during Kalman smoothing. This viewpoint encompasses most inference schemes, including expectation propagation (EP), the classical (Extended, Unscented, etc.) Kalman smoothers, and variational inference. We provide a unifying perspective on these algorithms, showing how replacing the power EP moment matching step with linearisation recovers the classical smoothers. EP provides some benefits over the traditional methods via introduction of the socalled cavity distribution, and we combine these benefits with the computational efficiency of linearisation, providing extensive empirical analysis demonstrating the efficacy of various algorithms under this unifying framework. We provide a fast implementation of all methods in JAX.

For agents

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

get_harvested_code_for_paper("2007.05994")
get_code_for_paper("2007.05994")
have("2007.05994")

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

curl https://syntology.ai/api/ran/2007.05994.json

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

[![Syntology run record](https://syntology.ai/api/ran/2007.05994.svg)](https://syntology.ai/paper/2007.05994)

Connect an agent — have() is free.