SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2102.06559 · 2021

Infinitely Deep Bayesian Neural Networks with Stochastic Differential Equations

arXiv · PDF · Open in the Atlas

Code that ran

3 of 6 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 0 have no contract verdict; 3 failures of our instrument; 3 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/2102.06559.json · badge: /api/ran/2102.06559.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
xwinxu/bayesde canonical (classifier label)
classifier basis: classifier_2025
3 ran · 0 no recorded run
copy not recorded community
not in the paper's declared repositories
0 ran · 3 no recorded run
FunctionStatusWhere it lives
ConcatLinear Ran
ran; our contract draft was wrong, not the code
xwinxu/bayesde/brax/_impl/diffeq_layers.py
code served (permissive licence) · get_code("ad88842e1ae6b0a4")
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 tuple, ndim 1, 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 jax
import jax.numpy as jnp
from jax import random
from jax.nn.initializers import he_normal, normal

# Set seed for reproducibility
key = random.PRNGKey(0)

# Create small test tensors
batch_size = 2
input_dim = 4
out_dim = 3

# Create dummy input: (x, t) tuple
# x shape: (batch_size, input_dim)
# t shape: (batch_size,) - time values
x = jnp.ones((batch_size, input_dim))
t = jnp.array([0.1, 0.2])

ARGS = {
    "out_dim": out_dim,
    "W_init": he_normal(),
    "b_init": normal(),
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
jax — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, 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
IgnoreLinear Ran
ran; our contract draft was wrong, not the code
xwinxu/bayesde/brax/_impl/diffeq_layers.py
code served (permissive licence) · get_code("7cebe4f80c9b8589")
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 tuple, ndim 1, 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 jax
import jax.numpy as jnp
from jax import random
from jax.nn.initializers import he_normal, normal

# Set seed for reproducibility
key = random.PRNGKey(0)

# Create initializers
W_init = he_normal()
b_init = normal()

# Define ARGS dict with all parameters
ARGS = {
    'out_dim': 5,
    'W_init': W_init,
    'b_init': b_init,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
jax — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, 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
shape_dependent Ran
ran; our contract draft was wrong, not the code
xwinxu/bayesde/brax/_impl/diffeq_layers.py
code served (permissive licence) · get_code("999e7c9766ea878a")
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 tuple, ndim 1, 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 numpy as np

torch.manual_seed(0)

# The function expects make_layer to be a callable that:
# 1. Takes an input_shape (tuple) as argument
# 2. Returns a tuple of (init_fun, apply_fun)
# 3. init_fun takes (rng, input_shape) and returns params
# 4. apply_fun takes (params, inputs, **kwargs) where inputs is a tuple (x, t)

def make_layer(input_shape):
    """
    A simple layer factory that creates a linear transformation layer.
    Takes input_shape and returns (init_fun, apply_fun) pair.
    """
    # input_shape will be something like (batch_size, feature_dim)
    
    def init_fun(rng, input_shape):
        # Initialize simple parameters (weights and bias)
        batch_size, feature_dim = input_shape
        # Return a dict of parameters
        params = {
            'weight': np.random.randn(feature_dim, 3),  # output dim = 3
            'bias': np.random.randn(3)
        }
        return params
    
    def apply_fun(params, inputs, **kwargs):
        # inputs is a tuple (x, t) where x is the spatial part and t is time
        x, t = inputs
        # Simple linear transformation on x
        x_np = x.numpy() if isinstance(x, torch.Tensor) else x
        output = np.dot(x_np, params['weight']) + params['bias']
        return torch.tensor(output, dtype=torch.float32)
    
    return (init_fun, apply_fun)


# Create test data
batch_size = 2
feature_dim = 4
time_dim = 3

# x: spatial input tensor
x = torch.randn(batch_size, feature_dim)

# t: time input tensor
t = torch.randn(batch_size, time_dim)

# inputs tuple as expected by apply_fun
inputs = (x, t)

# rng for initialization
rng = np.random.RandomState(0)

ARGS = {
    'make_layer': make_layer
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, 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
sdeint_ito Not yet run
unverified (harvested, no recorded run)
this paper's copy was not recorded; identical code first harvested from xwinxu/bayesde
pointer only · get_code("d82ed5ed65c58a23")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

sdeint_ito_fixed_grid Not yet run
unverified (harvested, no recorded run)
this paper's copy was not recorded; identical code first harvested from xwinxu/bayesde
pointer only · get_code("c67ae22fcae712b7")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

sdeint_strat Not yet run
unverified (harvested, no recorded run)
this paper's copy was not recorded; identical code first harvested from xwinxu/bayesde
pointer only · get_code("dd6e88e990764ff3")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

Also present in other papers' code

Of 6 samples lifted from this paper's linked repositories, 6 carry a function identity and 0 do not. Of the 6 distinct functions with an identity, 3 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 0 the same month, 0 later) and 3 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.

FunctionAlso present inEarliest-datedOthers named
sdeint_ito 1 other paper
1 dated earlier
2001.01328 (2020-01)
Scalable Gradients for Stochastic Differential Equations
sdeint_ito_fixed_grid 1 other paper
1 dated earlier
2001.01328 (2020-01)
Scalable Gradients for Stochastic Differential Equations
sdeint_strat 1 other paper
1 dated earlier
2001.01328 (2020-01)
Scalable Gradients for Stochastic Differential Equations

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 perform scalable approximate inference in continuous-depth Bayesian neural networks. In this model class, uncertainty about separate weights in each layer gives hidden units that follow a stochastic differential equation. We demonstrate gradient-based stochastic variational inference in this infinite-parameter setting, producing arbitrarily-flexible approximate posteriors. We also derive a novel gradient estimator that approaches zero variance as the approximate posterior over weights approaches the true posterior. This approach brings continuous-depth Bayesian neural nets to a competitive comparison against discrete-depth alternatives, while inheriting the memory-efficient training and tunable precision of Neural ODEs.

For agents

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

get_harvested_code_for_paper("2102.06559")
get_code_for_paper("2102.06559")
have("2102.06559")

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

curl https://syntology.ai/api/ran/2102.06559.json

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

[![Syntology run record](https://syntology.ai/api/ran/2102.06559.svg)](https://syntology.ai/paper/2102.06559)

Connect an agent — have() is free.