Papers › Self-Supervised Neuron Segmentation with Multi-Agent Reinforcement Learning

Self-Supervised Neuron Segmentation with Multi-Agent Reinforcement Learning

6 Oct 2023arXiv:2310.04148archive 2025-07-28

Yinda Chen, Wei Huang, Shenglong Zhou, Qi Chen, Zhiwei Xiong

The performance of existing supervised neuron segmentation methods is highly dependent on the number of accurate annotations, especially when applied to large scale electron microscopy (EM) data. By extracting semantic information from unlabeled data, self-supervised methods can improve the performance of downstream tasks, among which the mask image model (MIM) has been widely used due to its simplicity and effectiveness in recovering original information from masked images. However, due to the high degree of structural locality in EM images, as well as the existence of considerable noise, many voxels contain little discriminative information, making MIM pretraining inefficient on the neuron segmentation task. To overcome this challenge, we propose a decision-based MIM that utilizes reinforcement learning (RL) to automatically search for optimal image masking ratio and masking strategy. Due to the vast exploration space, using single-agent RL for voxel prediction is impractical. Therefore, we treat each input patch as an agent with a shared behavior policy, allowing for multi-agent collaboration. Furthermore, this multi-agent model can capture dependencies between voxels, which is beneficial for the downstream segmentation task. Experiments conducted on representative EM datasets demonstrate that our approach has a significant advantage over alternative self-supervised methods on the task of neuron segmentation. Code is available at https://github.com/ydchen0806/dbMiM.

PaperPDFCodeCode Syntology ranMethod as code

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 24 code samples harvested from 1 repository linked to this paper: 20 ran (of which 4 constructed an object rather than computing a result; 18 with no instrument failure: 2 honoured, 1 violated, 15 with no contract checked; 2 where Syntology's instrument failed) · 4 unverified. Where Syntology's instrument failed: 2 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 24 samples from 1 repository, 20 ran (18 with no instrument failure; 2 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

ydchen0806/dbmim officialmentioned in paperpytorchSyntology: 24 samples, 20 ran (2 instrument-failed) · 4 unverified report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

24 samples harvested: 20 ran (of which 4 constructed an object rather than computing a result; 18 with no instrument failure: 2 honoured, 1 violated, 15 with no contract checked; 2 where Syntology's instrument failed) · 4 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

2ran · honoured contract
1ran · violated contract
2ran · our draft was wrong
15ran
4unverified

Licence: 24 of the 24 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from ydchen0806/dbmim. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. For 2 samples that ran, the graph records a signature this page does not show: for a class the fixture built without calling, the harness called the object with inputs it made up itself, so the recorded signature is the harness's probe, not the fixture's arguments. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity.

ydchen0806/dbmim official repository 24 samples, 20 ran (2 instrument-failed) · 4 unverified
DecisionModule dbmim/models.py ran fingerprinted no licence file found · pointer only · b4ad2611c31e1f3c · report fixture: returned DecisionModule · level recorded 2026-09-22
MAEOutput dbmim/models.py ran no licence file found · pointer only · ffa0da027cebb013 · report fixture: returned MAEOutput · level recorded 2026-09-22
PatchEmbed3D in 7 papers dbmim/models.py ran no licence file found · pointer only · 7dfdb20a93a38877 · report fixture: returned PatchEmbed3D · level recorded 2026-09-22
TransformerBlock in 73 papers dbmim/models.py ran no licence file found · pointer only · a8c788e1c3eb20c1 · report fixture: returned TransformerBlock · level recorded 2026-09-22
_decision_needs_edge_scores dbmim/models.py ran · violated contract no licence file found · pointer only · 640410b6903d7f7a · report fixture: signature decision_module:callable/0/? · returned bool · level recorded 2026-09-22
_normalize_axis_weights dbmim/models.py ran · honoured contract fingerprinted no licence file found · pointer only · 0d706bb7ac1a82b9 · report fixture: signature axis_weights:NoneType/0/? · returned tuple · level recorded 2026-09-22
_triple dbmim/models.py ran · honoured contract fingerprinted no licence file found · pointer only · 17347b00ad3c515c · report fixture: signature value:int/0/int · returned tuple · level recorded 2026-09-22
affinities_to_connected_components dbmim/postprocess.py ran no licence file found · pointer only · cf5e36b4f5d5bafc · report fixture: signature affinities:ndarray/4/float, threshold:float/0/float, min_size:int/0/int · returned ndarray · level recorded 2026-09-24
augment_image_and_label dbmim/datasets.py ran no licence file found · pointer only · ffbb8cf276ef9f02 · report fixture: signature volume:Tensor/4/float, label:Tensor/4/float, rotate_xy:bool/0/bool, gamma:bool/0/bool, gamma_range:tuple/1/float, gamma_probability:float/0/float, noise_std:float/0/float, intensity_probability:float/0/float, intensity_gain_range:tuple/1/float, intensity_bias_range:tuple/1/float, xy_blur_probability:float/0/float, xy_blur_kernel:int/0/int · returned tuple · level recorded 2026-09-24
binary_iou_from_logits dbmim/metrics.py ran fingerprinted no licence file found · pointer only · d00a010268c413ef · report fixture: signature logits:Tensor/2/float, target:Tensor/2/float, threshold:float/0/float · returned float · level recorded 2026-09-24
device_from_config dbmim/utils.py ran no licence file found · pointer only · f0e1aff9aea1ad2d · report fixture: signature config:dict/0/? · returned device · level recorded 2026-09-24
dice_from_logits dbmim/metrics.py ran fingerprinted no licence file found · pointer only · a8c9b35dce17bae5 · report fixture: signature logits:Tensor/2/float, target:Tensor/2/float, threshold:float/0/float · returned float · level recorded 2026-09-24
ensure_dir in 22 papers dbmim/utils.py ran no licence file found · pointer only · 3f46792b39d12fa5 · report fixture: signature path:str/0/str · returned PosixPath · level recorded 2026-09-24
gradient_loss_3d dbmim/models.py ran · our draft was wrong fingerprinted no licence file found · pointer only · 8505f28e0bda17d4 · report fixture: signature pred:Tensor/4/float, target:Tensor/4/float, axis_weights:list/1/float · returned Tensor · level recorded 2026-09-22
is_main train_finetune.py ran fingerprinted no licence file found · pointer only · 02cf333b1f5f0dd4 · report fixture: signature rank:int/0/int · returned bool · level recorded 2026-09-24
load_config in 389 papers dbmim/utils.py ran no licence file found · pointer only · 52e93d4c7dc1aa9b · report fixture: signature path:str/0/str · returned dict · level recorded 2026-09-24
membrane_edge_map_3d dbmim/models.py ran · our draft was wrong fingerprinted no licence file found · pointer only · 3dce450f399d1158 · report fixture: signature x:Tensor/5/float, axis_weights:list/1/float, clip:float/0/float · returned Tensor · level recorded 2026-09-22
normalize_volume dbmim/datasets.py ran fingerprinted no licence file found · pointer only · 83684b4d0057676f · report fixture: signature volume:ndarray/3/int · returned ndarray · level recorded 2026-09-24
random_crop_3d dbmim/datasets.py ran no licence file found · pointer only · a2a94ab0f75b1a93 · report fixture: signature volume:ndarray/3/float, crop_size:tuple/1/int, rng:Random/0/? · returned ndarray · level recorded 2026-09-24
relabel_sequential dbmim/postprocess.py ran fingerprinted no licence file found · pointer only · d9188e9f5325493b · report fixture: signature labels:ndarray/1/int · returned ndarray · level recorded 2026-09-24
DBMIM3DMAE dbmim/models.py unverified no licence file found · pointer only · bbe3d482f52125db · report
load_pretrained_backbone dbmim/models.py unverified no licence file found · pointer only · 2c0453237d66c5cb · report
unwrap in 16 papers train_finetune.py unverified no licence file found · pointer only · ba4d5d62b9e2f2a3 · report
watershed_fragments dbmim/postprocess.py unverified no licence file found · pointer only · 925f14fbca20ff8d · report

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it.

1 assembly on this paper: 1 harvested, 0 generated.

harvested Assembly for decision-based MIM (graph method: decision-based MIM). 10 bricks; individually on a fixture: 9 ran (of which 4 constructed an object rather than computing a result; 7 with no instrument failure: 2 honoured, 1 violated, 4 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified. The assembly has not been run end-to-end.

  1. DBMIM3DMAE entry · level 0 · unverified · bbe3d482f52125db
  2. DecisionModule model · level 1 · ran · b4ad2611c31e1f3c
  3. MAEOutput loss · level 1 · ran · ffa0da027cebb013
  4. PatchEmbed3D model · level 1 · ran · 7dfdb20a93a38877
  5. TransformerBlock model · level 1 · ran · a8c788e1c3eb20c1
  6. _decision_needs_edge_scores helper · level 1 · ran · violated contract · 640410b6903d7f7a
  7. _normalize_axis_weights helper · level 1 · ran · honoured contract · 0d706bb7ac1a82b9
  8. _triple helper · level 1 · ran · honoured contract · 17347b00ad3c515c
  9. gradient_loss_3d helper · level 1 · ran · our draft was wrong · 8505f28e0bda17d4
  10. membrane_edge_map_3d helper · level 1 · ran · our draft was wrong · 3dce450f399d1158

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2310.04148") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Multi-agent Reinforcement LearningReinforcement LearningReinforcement Learning (RL)Segmentationreinforcement-learning

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

MIM

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections