Papers › Revisiting Global Pooling through the Lens of Optimal Transport

Revisiting Global Pooling through the Lens of Optimal Transport

23 Jan 2022arXiv:2201.09191archive 2025-07-28

Minjie Cheng, Hongteng Xu

Global pooling is one of the most significant operations in many machine learning models and tasks, whose implementation, however, is often empirical in practice. In this study, we develop a novel and solid global pooling framework through the lens of optimal transport. We demonstrate that most existing global pooling methods are equivalent to solving some specializations of an unbalanced optimal transport (UOT) problem. Making the parameters of the UOT problem learnable, we unify various global pooling methods in the same framework, and accordingly, propose a generalized global pooling layer called UOT-Pooling (UOTP) for neural networks. Besides implementing the UOTP layer based on the classic Sinkhorn-scaling algorithm, we design a new model architecture based on the Bregman ADMM algorithm, which has better numerical stability and can reproduce existing pooling layers more effectively. We test our UOTP layers in several application scenarios, including multi-instance learning, graph classification, and image classification. Our UOTP layers can either imitate conventional global pooling layers or learn some new pooling mechanisms leading to better performance.

PaperPDFCodeCode Syntology ran

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

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2201.09191")

Code

Syntology Ran 1 of 15 code samples harvested from 1 repository linked to this paper; 14 have no recorded run. Of those that ran: 1 ran · fixture could not drive it.

By repository: community (archive-listed): 15 samples from 1 repository, 1 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

sds-lab/rot-pooling mentioned on GitHubpytorchMIT 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

15 samples harvested; 1 ran; 0 honoured the contract we drafted; 14 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

1ran · fixture could not drive it
14unverified

Licence: 0 of the 15 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 sds-lab/rot-pooling. “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. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (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 from a repository marked official are listed first. 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 four 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; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). 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.

accuracy sds-lab/rot-pooling/backbones/resnet_imagenet.py community (archive-listed) ran · fixture could not drive it MIT (permissive) · 131a82fd65128218 · report
get_dim sds-lab/rot-pooling/backbones/attention_mil.py community (archive-listed) unverified MIT (permissive) · ec8c933c0c321597 · report
global_softmax_pooling sds-lab/rot-pooling/pooling/baselines.py community (archive-listed) unverified MIT (permissive) · 9d7c914f74cff214 · report
initialize_edge_weight sds-lab/rot-pooling/backbones/adgcl.py community (archive-listed) unverified MIT (permissive) · e5f5692f50afab40 · report
initialize_node_features sds-lab/rot-pooling/backbones/adgcl.py community (archive-listed) unverified MIT (permissive) · 4765da09c326c59f · report
load_mil_data_mat sds-lab/rot-pooling/backbones/attention_mil.py community (archive-listed) unverified MIT (permissive) · a919d94d33818c15 · report
org_batch sds-lab/rot-pooling/backbones/ddi_gin.py community (archive-listed) unverified MIT (permissive) · 9a2e477393a1dffe · report
rot_sinkhorn sds-lab/rot-pooling/pooling/rotlayers.py community (archive-listed) unverified MIT (permissive) · 72b6375204b98c11 · report
set_tu_dataset_y_shape sds-lab/rot-pooling/backbones/adgcl.py community (archive-listed) unverified MIT (permissive) · 382c92abf2446b95 · report
to_sparse_batch sds-lab/rot-pooling/pooling/rotpooling.py community (archive-listed) unverified MIT (permissive) · 1d61626dd637955d · report
train sds-lab/rot-pooling/backbones/attention_mil.py community (archive-listed) unverified MIT (permissive) · 3ce20c8e0eb7e9c5 · report
train sds-lab/rot-pooling/backbones/ddi_gin.py community (archive-listed) unverified MIT (permissive) · 954a0c90bea1cc24 · report
uot_badmm sds-lab/rot-pooling/pooling/rotlayers.py community (archive-listed) unverified MIT (permissive) · 8ab6d9d56daf6d67 · report
uot_sinkhorn sds-lab/rot-pooling/pooling/rotlayers.py community (archive-listed) unverified MIT (permissive) · f3ed0e56606b76fe · report
validate sds-lab/rot-pooling/backbones/resnet_imagenet.py community (archive-listed) unverified MIT (permissive) · f7cef679653b52d5 · report

Tasks

Graph ClassificationGraph EmbeddingImage Classificationimage-classification

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

ADMM

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