Zobrazeno 1 - 10
of 22
pro vyhledávání: '"Martinkus, Karolis"'
In the realm of generative models for graphs, extensive research has been conducted. However, most existing methods struggle with large graphs due to the complexity of representing the entire joint distribution across all node pairs and capturing bot
Externí odkaz:
http://arxiv.org/abs/2312.11529
Autor:
Martinkus, Karolis, Ludwiczak, Jan, Cho, Kyunghyun, Liang, Wei-Ching, Lafrance-Vanasse, Julien, Hotzel, Isidro, Rajpal, Arvind, Wu, Yan, Bonneau, Richard, Gligorijevic, Vladimir, Loukas, Andreas
We introduce AbDiffuser, an equivariant and physics-informed diffusion model for the joint generation of antibody 3D structures and sequences. AbDiffuser is built on top of a new representation of protein structure, relies on a novel architecture for
Externí odkaz:
http://arxiv.org/abs/2308.05027
We provide a novel approach to construct generative models for graphs. Instead of using the traditional probabilistic models or deep generative models, we propose to instead find an algorithm that generates the data. We achieve this using evolutionar
Externí odkaz:
http://arxiv.org/abs/2304.12895
Rigid origami has shown potential in large diversity of practical applications. However, current rigid origami crease pattern design mostly relies on known tessellations. This strongly limits the diversity and novelty of patterns that can be created.
Externí odkaz:
http://arxiv.org/abs/2211.13219
Denoising diffusion probabilistic models and score-matching models have proven to be very powerful for generative tasks. While these approaches have also been applied to the generation of discrete graphs, they have, so far, relied on continuous Gauss
Externí odkaz:
http://arxiv.org/abs/2210.01549
We present a novel graph neural network we call AgentNet, which is designed specifically for graph-level tasks. AgentNet is inspired by sublinear algorithms, featuring a computational complexity that is independent of the graph size. The architecture
Externí odkaz:
http://arxiv.org/abs/2206.11010
We propose the fully explainable Decision Tree Graph Neural Network (DT+GNN) architecture. In contrast to existing black-box GNNs and post-hoc explanation methods, the reasoning of DT+GNN can be inspected at every step. To achieve this, we first cons
Externí odkaz:
http://arxiv.org/abs/2205.13234
We approach the graph generation problem from a spectral perspective by first generating the dominant parts of the graph Laplacian spectrum and then building a graph matching these eigenvalues and eigenvectors. Spectral conditioning allows for direct
Externí odkaz:
http://arxiv.org/abs/2204.01613
This paper studies Dropout Graph Neural Networks (DropGNNs), a new approach that aims to overcome the limitations of standard GNN frameworks. In DropGNNs, we execute multiple runs of a GNN on the input graph, with some of the nodes randomly and indep
Externí odkaz:
http://arxiv.org/abs/2111.06283
Different studies of the embedding space of transformer models suggest that the distribution of contextual representations is highly anisotropic - the embeddings are distributed in a narrow cone. Meanwhile, static word representations (e.g., Word2Vec
Externí odkaz:
http://arxiv.org/abs/2109.13304