Introduces graph-to-image prediction of per-node dynamic stability landscapes in oscillator networks from topology, releases two 10k-graph datasets, and shows GNN-CNN models achieve good accuracy with cross-size generalization.
hub Mixed citations
Relational inductive biases, deep learning, and graph networks
Mixed citation behavior. Most common role is background (56%).
67
Pith papers citing it
Background
56% of classified citations
abstract
Artificial intelligence (AI) has undergone a renaissance recently, making major progress in key domains such as vision, language, control, and decision-making. This has been due, in part, to cheap data and cheap compute resources, which have fit the natural strengths of deep learning. However, many defining characteristics of human intelligence, which developed under much different pressures, remain out of reach for current approaches. In particular, generalizing beyond one's experiences--a hallmark of human intelligence from infancy--remains a formidable challenge for modern AI. The following is part position paper, part review, and part unification. We argue that combinatorial generalization must be a top priority for AI to achieve human-like abilities, and that structured representations and computations are key to realizing this objective. Just as biology uses nature and nurture cooperatively, we reject the false choice between "hand-engineering" and "end-to-end" learning, and instead advocate for an approach which benefits from their complementary strengths. We explore how using relational inductive biases within deep learning architectures can facilitate learning about entities, relations, and rules for composing them. We present a new building block for the AI toolkit with a strong relational inductive bias--the graph network--which generalizes and extends various approaches for neural networks that operate on graphs, and provides a straightforward interface for manipulating structured knowledge and producing structured behaviors. We discuss how graph networks can support relational reasoning and combinatorial generalization, laying the foundation for more sophisticated, interpretable, and flexible patterns of reasoning. As a companion to this paper, we have released an open-source software library for building graph networks, with demonstrations of how to use them in practice.
hub tools
citation-role summary
background 8
baseline 1
citation-polarity summary
representative citing papers
Introduces transitive inference with exceptions task and analytically shows kernel ridge regression balances relational generalization and memorization depending on representational geometry, with validation in finetuned language models.
GraphScan replaces geometric or coordinate-based scanning in Vision SSMs with learned local semantic graph routing, yielding SOTA results among such models on classification and segmentation tasks.
Graph neural networks can approximate full 3D non-LTE Ca II populations in solar models with correlations above 0.99 and extreme computational efficiency.
The paper unifies emerging graph-based world models under a new paradigm and proposes a taxonomy organized by spatial, physical, and logical relational inductive biases.
PiGGO integrates a learned graph neural ODE as the continuous-time dynamics model within an extended Kalman filter to enable online virtual sensing and uncertainty-aware state estimation for nonlinear dynamic systems with unknown model form and sparse sensing.
Scale-autoregressive modeling (SAR) samples fluid flow distributions hierarchically from coarse to fine resolutions on meshes, achieving lower distributional error and 2-7x faster runtime than diffusion or flow-matching baselines.
A self-supervised multimodal alignment step plus equivariant GNN-based MARL yields over twofold sensing accuracy and 50% performance gains in decentralized V2I rate maximization.
Smoothness assumptions on graphical model kernels produce Wasserstein estimation rates determined by local graph structure rather than ambient dimension.
Causal Process Models reframe dynamic causal graph discovery as multi-agent reinforcement learning to build sparse time-varying graphs only at active interactions, outperforming dense baselines on physical prediction.
In-weights learning induces linear embeddings enabling transitive inference in transformers, whereas in-context learning defaults to match-and-copy unless pre-trained on linear tasks or prompted with linear mental maps.
Unsupervised GNN model learns local updates for approximate MaxIS on dynamic graphs, achieving competitive ratios on 200-1000 node instances and 1.00-1.18x larger solutions than other unsupervised models when generalizing to 100x larger graphs.
Temporal Graph Networks combine memory modules and graph operators to learn on dynamic graphs as timed event sequences, outperforming prior methods on transductive and inductive tasks while unifying earlier models as special cases.
Graph Kernel Networks learn PDE solution operators that generalize across discretization methods and grid resolutions using graph-based kernel integration.
BERT stores relational knowledge extractable via cloze queries without fine-tuning and matches supervised baselines on open-domain QA tasks.
Placeto learns generalizable RL policies for device placement via iterative improvements and graph embeddings, needing up to 6.1x fewer steps than prior methods and applying to unseen graphs without retraining.
PyTorch Geometric is a PyTorch library that delivers fast graph neural network training through sparse GPU kernels and variable-size mini-batching.
A graph neural network learns to approximate altruistic robot transfers across heterogeneous teams using Hamilton's rule, achieving near-optimal allocation in simulated firefighting scenarios.
GOAL uses conditioned diffusion on relational graphs with typed edges to produce feasible multi-objective solutions for scheduling problems, reporting 100% feasibility and sub-0.2% MAPE on FSP, JSP, and FJSP up to 20 jobs.
Ada-Diffuser is a causal diffusion model that jointly learns observed interaction structure and underlying latent dynamics from minimal observations for adaptive planning and policy learning.
Neural point-forms are introduced as permutation-invariant neural layers that output learned form-comparison matrices for point clouds, with a claimed consistency proof under sampling and manifold assumptions and competitive results on synthetic and biological data.
SACHI enriches agent representations via graph transformer convolutions over inter-agent graphs to enable holistic information integration, outperforming baselines across five cooperative tasks with statistical significance.
LINC decouples local consequence scoring from hidden matching in constructive neural routing solvers, cutting CVRPTW gaps for PolyNet from 13.83%/38.15% to 7.26%/14.71% on Solomon/Homberger benchmarks.
Sheet as Token represents each worksheet as a single dense token and uses a multi-channel graph retriever to improve retrieval of supporting sheets in multi-sheet workbooks.