SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · aaai_20764 · AAAI · 2022

Blockwise Sequential Model Learning for Partially Observable Reinforcement Learning

Giseung Park, Youngchul Sung, Sungho Choi

Code that ran

8 of 9 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 7 have no contract verdict; 1 failure of our instrument; 1 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/aaai_20764.json · badge: /api/ran/aaai_20764.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
Giseung-Park/BlockSeq canonical (classifier label)
classifier basis: census
8 ran · 1 no recorded run
FunctionStatusWhere it lives
dot_scaled_attention Ran
ran on a synthesized input
Giseung-Park/BlockSeq/code_off_policy/attention_model.py
code served (permissive licence) · get_code("be450c69f851a689")
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. 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 math

torch.manual_seed(0)

# Create small tensors with distinct dimensions
# sequence_length=3, batch_size=2, d_k=4
seq_len = 3
batch_size = 2
d_k = 4

query = torch.randn(seq_len, batch_size, d_k)
key = torch.randn(seq_len, batch_size, d_k)
value = torch.randn(seq_len, batch_size, d_k)

ARGS = {
    "query": query,
    "key": key,
    "value": value,
}
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
get_model_dir Ran
ran on a synthesized input
Giseung-Park/BlockSeq/code_on_policy_minigrid/utils/storage.py
code served (permissive licence) · get_code("5c010a9d40b9931b")
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 str. 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 os

# The function get_model_dir takes a model_name parameter and constructs a path
# by joining the storage directory with the model name.
# It uses os.path.join() which expects string arguments.

# Set up environment - we can optionally set RL_STORAGE
os.environ['RL_STORAGE'] = 'storage'

# Define ARGS with the model_name parameter
ARGS = {
    'model_name': 'my_model'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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_status_path Ran
ran on a synthesized input
Giseung-Park/BlockSeq/code_on_policy_minigrid/utils/storage.py
code served (permissive licence) · get_code("7312b91ca25835e3")
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 str. 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 os
import tempfile

# Create a temporary directory to use as model_dir
temp_dir = tempfile.mkdtemp()

ARGS = {
    'model_dir': temp_dir
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
moving_average Ran
ran; our contract draft was wrong, not the code
Giseung-Park/BlockSeq/code_off_policy/plotter.py
pointer only (licence: MIT); the stored copy get_code serves, from MaryamAlipourH/prob-unet-climate-downscaling, is pointer only (licence: NONE) · get_code("265136f0d46b4448")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from MaryamAlipourH/prob-unet-climate-downscaling, and GitHub detected no licence in that repository when it was checked. 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. A run is recorded.

preprocess_images Ran
ran on a synthesized input
Giseung-Park/BlockSeq/code_on_policy_minigrid/utils/format.py
code served (permissive licence) · get_code("aae66f156a5e4829")
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 3, dtype family float, shape (2, 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 numpy
import torch

torch.manual_seed(0)

# The function takes images and converts them to a torch tensor
# images can be a list, numpy array, or other array-like structure
# device is optional and defaults to None

# Create a small batch of images (batch_size=2, height=4, width=3, channels=1)
images = [
    [[1.0, 2.0, 3.0], [4.0, 5.0, 6.0], [7.0, 8.0, 9.0], [10.0, 11.0, 12.0]],
    [[13.0, 14.0, 15.0], [16.0, 17.0, 18.0], [19.0, 20.0, 21.0], [22.0, 23.0, 24.0]]
]

ARGS = {
    "images": images,
    "device": None
}
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 3, dtype family float, shape (2, 4, 3) — shape (2, 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
preprocess_texts Ran
ran on a synthesized input
Giseung-Park/BlockSeq/code_on_policy_minigrid/utils/format.py
code served (permissive licence) · get_code("4f10e49463dc013c")
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 int, 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
import numpy
import re

torch.manual_seed(0)

# Build a small vocabulary mapping tokens to indices
vocab = {
    'hello': 0,
    'world': 1,
    'foo': 2,
    'bar': 3,
    'test': 4,
    'python': 5
}

# Build sample texts to preprocess
texts = [
    'hello world',
    'foo bar test python'
]

# Set device to None (CPU)
device = None

ARGS = {
    'texts': texts,
    'vocab': vocab,
    'device': device
}
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 2, dtype family int, 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
synthesize Ran
ran on a synthesized input
Giseung-Park/BlockSeq/code_on_policy_minigrid/utils/other.py
code served (permissive licence) · get_code("8a213cad834c4156")
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 OrderedDict. 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 numpy
import collections

# The function 'synthesize' takes an 'array' parameter
# and computes mean, std, min, max using numpy functions.
# So we need to pass a numpy array.

array = numpy.array([1.0, 2.0, 3.0, 4.0, 5.0])

ARGS = {
    'array': array
}
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 OrderedDict — a OrderedDict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a OrderedDict 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
xy2updatenumy Ran
ran on a synthesized input
Giseung-Park/BlockSeq/code_off_policy/plotter.py
code served (permissive licence) · get_code("51d266a0e240593e")
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 2, 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

# Analyze the function:
# - x: numpy array, used with np.where(x < ...) comparison and indexing
# - y: numpy array, indexed with ind[...] where ind comes from np.where
# - update_timestep: numeric value, used in arithmetic (division, multiplication)
# - max_timestep: numeric value, used in arithmetic (division, floor)

# Create sample data
# x should be a 1D array of timestep values (increasing)
x = np.array([0.1, 0.5, 1.2, 1.8, 2.5, 3.1])

# y should be a 1D array of the same length as x (values to pick from)
y = np.array([10.0, 20.0, 30.0, 40.0, 50.0, 60.0])

# Scalar parameters for time stepping
update_timestep = 1.0
max_timestep = 3.0

ARGS = {
    'x': x,
    'y': y,
    'update_timestep': update_timestep,
    'max_timestep': max_timestep
}
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 tuple, ndim 2, 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
get_status Not yet run
unverified (harvested, no recorded run)
Giseung-Park/BlockSeq/code_on_policy_minigrid/utils/storage.py
code served (permissive licence) · get_code("c7e7590a49d1d1c7")
How it ran

no invocation recorded.

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 distinct functions with an identity, 2 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 1 later) and 7 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
moving_average 3 other papers
1 dated earlier, 2 later
1804.05394 (2018-04)
Deep optimal stopping
2404.15274 (2024-04)
Metric-Guided Conformal Bounds for Probabilistic Image Reconstruction
2511.03197 (2025-11)
A Probabilistic U-Net Approach to Downscaling Climate Simulations
preprocess_images 1 other paper
1 same year, order not determinable
2211.10282 (2022-11)
Exploring through Random Curiosity with General Value Functions

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

This paper proposes a new sequential model learning architecture to solve partially observable Markov decision problems. Rather than compressing sequential information at every timestep as in conventional recurrent neural network-based methods, the proposed architecture generates a latent variable in each data block with a length of multiple timesteps and passes the most relevant information to the next block for policy optimization. The proposed blockwise sequential model is implemented based on self-attention, making the model capable of detailed sequential learning in partial observable settings. The proposed model builds an additional learning network to efficiently implement gradient estimation by using self-normalized importance sampling, which does not require the complex blockwise input data reconstruction in the model learning. Numerical results show that the proposed method significantly outperforms previous methods in various partially observable environments.

For agents

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

get_harvested_code_for_paper("aaai_20764")
get_code_for_paper("aaai_20764")
have("aaai_20764")

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

curl https://syntology.ai/api/ran/aaai_20764.json

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

[![Syntology run record](https://syntology.ai/api/ran/aaai_20764.svg)](https://syntology.ai/paper/aaai_20764)

Connect an agent — have() is free.