eLife

A differentiable model for optimizing the genetic drivers of synaptogenesis

This study presents SynaptoGen, a differentiable extension of connectome models that links gene expression, protein-protein interaction probabilities, synaptic multiplicity, and synaptic weights, and ...
GitHub

DSG-JIT: A JIT‑Compiled, Differentiable 3D Dynamic Scene Graph Engine

Modern spatial intelligence systems—SLAM pipelines, neural rendering models, and 3D scene graph frameworks—remain fragmented. Each solves part of the perception problem, but none unify: Metric ...
IEEE

Differentiable Graph Centrality

Abstract: We view graph centrality algorithms as differentiable processes and explore the implications of this lens.First, we revisit PageRank, an ubiquitous graph centrality algorithm, and consider ...
Frontiers

Orthogonality and graph divergence losses promote disentanglement in generative models

Over the last decade, deep generative models have evolved to generate realistic and sharp images. The success of these models is often attributed to an extremely large number of trainable parameters ...
marktechpost

Enhancing Graph Classification with Edge-Node Attention-based Differentiable Pooling and Multi-Distance Graph Neural Networks GNNs

Graph Neural Networks GNNs are advanced tools for graph classification, leveraging neighborhood aggregation to update node representations iteratively. This process captures local and global graph ...
marktechpost

This AI Paper from China Introduces SGTR and SGTR+: An End-to-End CNN-Transformer-Based Scene Graph Generating Framework and Its Extension

Understanding the intricate relationships between visual entities in a scene is pivotal for scene comprehension. These relationships can be expressed as triplets, forming a scene graph with entities ...
Scientific Research Publishing

Ying, Z.T., You, J.X., Morris, C., Ren, X., Hamilton, W. and Leskovec, J. (2018) Hierarchical Graph Representation Learning with Differentiable Pooling. ArXiv: 1806.08804.

ABSTRACT: Node of network has lots of information, such as topology, text and label information. Therefore, node classification is an open issue. Recently, one vector of node is directly connected at ...
IEEE

NeuralPUMA: Learning to Phase Unwrap Through Differentiable Graph Cuts

Abstract: Deep learning solutions have recently demonstrated remarkable performance in phase unwrapping by approaching the problem as a semantic segmentation task. However, these solutions lack ...