</> Code with Papers by Syntology

Methods › Graphs › Graph Models › MPNN

Message Passing Neural Network

MPNN

74 papers tagged archive 2025-07-28

Introduced by Justin Gilmer et al. in Neural Message Passing for Quantum Chemistry

archive 2025-07-28 Description, source and code snippet are the archive's method entry.

There are at least eight notable examples of models from the literature that can be described using the Message Passing Neural Networks (MPNN) framework. For simplicity we describe MPNNs which operate on undirected graphs G with node features xᵥ and edge features e_(vw). It is trivial to extend the formalism to directed multigraphs. The forward pass has two phases, a message passing phase and a readout phase. The message passing phase runs for T time steps and is defined in terms of message functions Mₜ and vertex update functions Uₜ. During the message passing phase, hidden states hᵥᵗ at each node in the graph are updated based on messages mᵥᵗ⁺¹ according to mᵥᵗ⁺¹ = ∑_(w ∈N(v)) Mₜ(hᵥᵗ, h_wᵗ, e_(vw)) hᵥᵗ⁺¹ = Uₜ(hᵥᵗ, mᵥᵗ⁺¹) where in the sum, N(v) denotes the neighbors of v in graph G. The readout phase computes a feature vector for the whole graph using some readout function R according to ŷ = R( hᵥᵀ | v ∈G ) The message functions Mₜ, vertex update functions Uₜ, and readout function R are all learned differentiable functions. R operates on the set of node states and must be invariant to permutations of the node states in order for the MPNN to be invariant to graph isomorphism.

PaperSource

Papers archive 2025-07-28

30 shown of 74, newest first. Repository counts are the archive's code-links table. A Syntology line states what Syntology ran from that paper's harvested code; it is per sample and not a correctness claim.

Tasks archive 2025-07-28

20 shown of 63 tasks the archive attaches to papers tagged with this method, by distinct papers. A task without a page in the catalog is plain text.

TaskPapers
Graph Neural Network16
Node Classification11
Graph Classification8
Prediction8
Graph Regression7
Link Prediction6
Property Prediction6
Formation Energy5
Graph Learning4
Inductive Bias4
Molecular Property Prediction4
Band Gap3
Benchmarking3
Drug Discovery3
Neural Architecture Search3
Classification2
Deep Learning2
Graph Property Prediction2
Graph Representation Learning2
Management2

Usage over time archive 2025-07-28

Papers per year tagged with MPNN: 2017 to 2025, peak 18 18 0 2017: 1 paper 2017 2018: 0 papers 2018 2019: 1 paper 2019 2020: 10 papers 2020 2021: 9 papers 2021 2022: 12 papers 2022 2023: 18 papers 2023 2024: 13 papers 2024 2025: 10 papers 2025
Papers per year the archive tags with this method, by the paper's archive date (74 dated). Bars are counts, not a trend claim.

Components: the archive holds no method-to-method composition, so PwC's Components table cannot be rebuilt; the Papers list carries no Results column for the same reason (the archive does not join its leaderboard rows to method tags).

Categories archive 2025-07-28

Graph Models

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