SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2402.10236 · 2024

Discovering Sensorimotor Agency in Cellular Automata using Diversity Search

arXiv · PDF · Open in the Atlas

Code that ran

4 of 4 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, 2 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/2402.10236.json · badge: /api/ran/2402.10236.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
flowersteam/sensorimotor-lenia-search canonical (classifier label)
classifier basis: census
4 ran · 0 no recorded run
FunctionStatusWhere it lives
is_long_term_stable Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
flowersteam/sensorimotor-lenia-search/expe/calc_categories.py
code served (permissive licence) · get_code("6e6e565873cee454")
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 bool, 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 numpy as np

# Create statistics dict with activation_mass
# The function expects activation_mass to be a numpy array with shape (n_steps,)
# We'll create a small array with 100 steps to have enough data points
# for the phase calculations to work meaningfully

activation_mass = np.array([
    0.5, 0.6, 0.55, 0.65, 0.58,  # phase 1 (indices 0-24, avg ~0.58)
    0.7, 0.65, 0.72, 0.68, 0.7,
    0.75, 0.72, 0.78, 0.76, 0.74,
    0.8, 0.77, 0.82, 0.79, 0.81,
    0.85, 0.83, 0.87, 0.86, 0.84,
    1.2, 1.15, 1.25, 1.18, 1.22,  # phase 2 (indices 75-99, avg ~1.2)
    1.3, 1.28, 1.32, 1.29, 1.31,
    1.35, 1.33, 1.38, 1.36, 1.34,
    1.4, 1.38, 1.42, 1.39, 1.41,
    1.45, 1.43, 1.48, 1.46, 1.44,
    0.55, 0.6, 0.58, 0.62, 0.59,
    0.65, 0.63, 0.68, 0.66, 0.64,
    0.7, 0.68, 0.73, 0.71, 0.69,
    0.75, 0.73, 0.78, 0.76, 0.74,
    0.8, 0.78, 0.83, 0.81, 0.79,
    1.25, 1.2, 1.3, 1.22, 1.28,
    1.35, 1.3, 1.4, 1.32, 1.38,
    1.45, 1.4, 1.5, 1.42, 1.48,
    1.55, 1.5, 1.6, 1.52, 1.58,
    1.65, 1.6, 1.7, 1.62, 1.68,
])

statistics = {
    'activation_mass': activation_mass
}

ARGS = {
    'statistics': statistics,
    'phase1_timepoints': (1 / 4, 1 / 2),
    'phase2_timepoints': (3 / 4, 1),
    'low_ratio_threshold': 0.5,
    'high_ratio_threshold': 3.0,
}
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 bool, 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
is_robust Ran
ran, violated the contract we drafted
flowersteam/sensorimotor-lenia-search/expe/calc_categories.py
code served (permissive licence) · get_code("5d39bb5a945c5caa")
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 bool, 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 torch
import torch.nn as nn

torch.manual_seed(0)

# The function accesses statistics['activation_mass'][-1].item()
# So we need a dict with 'activation_mass' key containing a list of tensors
# The last element should be a tensor with .item() method (scalar tensor)

activation_mass_list = [
    torch.tensor(0.5),
    torch.tensor(0.6),
    torch.tensor(0.3)
]

statistics = {
    'activation_mass': activation_mass_list
}

ARGS = {
    'statistics': statistics,
    'low_mass_threshold': 0,
    'high_mass_threshold': 6400,
    'num_cells': 65536
}
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 bool, 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
is_soliton Ran
ran, violated the contract we drafted
flowersteam/sensorimotor-lenia-search/expe/calc_categories.py
code served (permissive licence) · get_code("145e76d51b285aff")
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 bool, 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 torch
import numpy as np

torch.manual_seed(0)

# The function expects:
# - statistics: a dict with keys 'connected_components_activity' and 'connected_components_nr_objects'
# - low_nr_objects: int, default 0
# - high_nr_objects: int, default 2
# - max_activity: float, default 256 * 256 * 0.6

# Create a statistics dict with the required keys
# activity should be a tensor with a .sum() method
# nr_objects should be an int or comparable value

activity_tensor = torch.tensor([10.0, 20.0, 15.0])
nr_objects_value = 1

statistics = {
    'connected_components_activity': activity_tensor,
    'connected_components_nr_objects': nr_objects_value
}

ARGS = {
    'statistics': statistics,
    'low_nr_objects': 0,
    'high_nr_objects': 2,
    'max_activity': 256 * 256 * 0.6
}
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 bool, 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
rle2arr Ran
ran, honoured the contract we drafted
flowersteam/sensorimotor-lenia-search/expe/handmade_crea_json2pickle.py
code served (permissive licence) · get_code("54be16d8c9455522")
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 (3, 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 torch
import re

torch.manual_seed(0)

# The function takes an RLE (Run Length Encoding) string as input
# Let's create a simple valid RLE string
# RLE format: numbers followed by characters (., b, o, or letters A-X)
# $ denotes end of row, ! denotes end of pattern
# Example: "3b2o$2bo$b2o!" means:
# - 3 dead cells (b or .), 2 live cells (o)
# - newline
# - 2 dead cells, 1 live cell
# - newline
# - 1 dead cell, 2 live cells

rle_string = "3b2o$2bo$b2o!"

ARGS = {
    "st": rle_string
}
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 (3, 5) — shape (3, 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

Also present in other papers' code

Of 4 samples lifted from this paper's linked repositories, 4 carry a function identity and 0 do not. Of the 4 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 4 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

The research field of Artificial Life studies how life-like phenomena such as autopoiesis, agency, or self-regulation can self-organize in computer simulations. In cellular automata (CA), a key open-question has been whether it it is possible to find environment rules that self-organize robust "individuals" from an initial state with no prior existence of things like "bodies", "brain", "perception" or "action". In this paper, we leverage recent advances in machine learning, combining algorithms for diversity search, curriculum learning and gradient descent, to automate the search of such "individuals", i.e. localized structures that move around with the ability to react in a coherent manner to external obstacles and maintain their integrity, hence primitive forms of sensorimotor agency. We show that this approach enables to find systematically environmental conditions in CA leading to self-organization of such basic forms of agency. Through multiple experiments, we show that the discovered agents have surprisingly robust capabilities to move, maintain their body integrity and navigate among various obstacles. They also show strong generalization abilities, with robustness to changes of scale, random updates or perturbations from the environment not seen during training. We discuss how this approach opens new perspectives in AI and synthetic bioengineering.

For agents

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

get_harvested_code_for_paper("2402.10236")
get_code_for_paper("2402.10236")
have("2402.10236")

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

curl https://syntology.ai/api/ran/2402.10236.json

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

[![Syntology run record](https://syntology.ai/api/ran/2402.10236.svg)](https://syntology.ai/paper/2402.10236)

Connect an agent — have() is free.