SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2010.09170 · 2020

Belief-Grounded Networks for Accelerated Robot Learning under Partial Observability

arXiv · PDF · Open in the Atlas

Code that ran

2 of 2 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; 1 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 1 failure 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/2010.09170.json · badge: /api/ran/2010.09170.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
hai-h-nguyen/belief-grounded-network canonical (classifier label)
classifier basis: census
2 ran · 0 no recorded run
FunctionStatusWhere it lives
get_monitor_files Ran
ran; our contract draft was wrong, not the code
hai-h-nguyen/belief-grounded-network/modifications/monitor.py
code served (permissive licence) · get_code("dc7cbce302df4a1e")
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 list, 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 tempfile
import os

# Create a temporary directory for monitoring files
temp_dir = tempfile.mkdtemp()

# Create some dummy monitor files in the directory
monitor_file1 = os.path.join(temp_dir, 'env1.monitor.csv')
monitor_file2 = os.path.join(temp_dir, 'env2.monitor.csv')

# Write minimal valid monitor files
with open(monitor_file1, 'w') as f:
    f.write('# {"t_start": 0, "env_id": "Test-v0"}\n')
    f.write('r,l,t\n')

with open(monitor_file2, 'w') as f:
    f.write('# {"t_start": 0, "env_id": "Test-v0"}\n')
    f.write('r,l,t\n')

ARGS = {
    'dir': temp_dir
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
gym — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, 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
load_results Ran
ran, honoured the contract we drafted
hai-h-nguyen/belief-grounded-network/modifications/monitor.py
code served (permissive licence) · get_code("ab9306edd1184fa8")
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 DataFrame. 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 tempfile
import os
import csv
import json

# Create a temporary directory for monitor files
tmpdir = tempfile.mkdtemp()

# Create a sample monitor.csv file
monitor_csv_path = os.path.join(tmpdir, 'test.monitor.csv')
with open(monitor_csv_path, 'wt') as f:
    f.write('# {"t_start": 0.0, "env_id": "CartPole-v1"}\n')
    f.write('r,l,t\n')
    f.write('100.5,50,10.5\n')
    f.write('95.2,48,20.3\n')

# Create a sample monitor.json file
monitor_json_path = os.path.join(tmpdir, 'test2.monitor.json')
with open(monitor_json_path, 'wt') as f:
    f.write('{"t_start": 0.0, "env_id": "CartPole-v1"}\n')
    f.write('{"r": 105.0, "l": 52, "t": 15.2}\n')
    f.write('{"r": 98.5, "l": 49, "t": 25.1}\n')

ARGS = {
    'dir': tmpdir
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
gym, pandas — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DataFrame — a DataFrame output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataFrame output the recorded ndim 2 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

Also present in other papers' code

Of 2 samples lifted from this paper's linked repositories, 2 carry a function identity and 0 do not. Of the 2 distinct functions with an identity, 2 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 2 later) and 0 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
get_monitor_files 4 other papers
3 dated earlier, 1 later
1811.02073 (2018-11)
QUOTA: The Quantile Option Architecture for Reinforcement Learning
1905.02363 (2019-05)
Dimension-Wise Importance Sampling Weight Clipping for Sample-Efficient Reinforcement Learning
2006.04061 (2020-06)
Dual Policy Distillation
and 1 more not named here
load_results 3 other papers
2 dated earlier, 1 later
1811.02073 (2018-11)
QUOTA: The Quantile Option Architecture for Reinforcement Learning
1905.02363 (2019-05)
Dimension-Wise Importance Sampling Weight Clipping for Sample-Efficient Reinforcement Learning
2605.18591 (2026-05)
Randomized Advantage Transformation (RAT): Computing Natural Policy Gradients via Direct Backpropagation

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

Many important robotics problems are partially observable in the sense that a single visual or force-feedback measurement is insufficient to reconstruct the state. Standard approaches involve learning a policy over beliefs or observation-action histories. However, both of these have drawbacks; it is expensive to track the belief online, and it is hard to learn policies directly over histories. We propose a method for policy learning under partial observability called the Belief-Grounded Network (BGN) in which an auxiliary belief-reconstruction loss incentivizes a neural network to concisely summarize its input history. Since the resulting policy is a function of the history rather than the belief, it can be executed easily at runtime. We compare BGN against several baselines on classic benchmark tasks as well as three novel robotic touch-sensing tasks. BGN outperforms all other tested methods and its learned policies work well when transferred onto a physical robot.

For agents

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

get_harvested_code_for_paper("2010.09170")
get_code_for_paper("2010.09170")
have("2010.09170")

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

curl https://syntology.ai/api/ran/2010.09170.json

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

[![Syntology run record](https://syntology.ai/api/ran/2010.09170.svg)](https://syntology.ai/paper/2010.09170)

Connect an agent — have() is free.