Results 11 to 20 of about 21,280,601 (295)
Atomistic structure learning [PDF]
The Journal of Chemical Physics, 2019 One endeavour of modern physical chemistry is to use bottom-up approaches to design materials and drugs with desired properties. Here we introduce an atomistic structure learning algorithm (ASLA) that utilizes a convolutional neural network to build 2D ...
Hammer, Bjørk +6 more
core +5 more sources
Learning hierarchically-structured concepts [PDF]
Neural Networks, 2021 We study the question of how concepts that have structure get represented in the brain. Specifically, we introduce a model for hierarchically structured concepts and we show how a biologically plausible neural network can recognize these concepts, and how it can learn them in the first place.
Nancy Lynch, Frederik Mallmann-Trenn
openaire +4 more sources
ESL-SNNs: An Evolutionary Structure Learning Strategy for Spiking Neural Networks [PDF]
AAAI Conference on Artificial Intelligence, 2023 Spiking neural networks (SNNs) have manifested remarkable advantages in power consumption and event-driven property during the inference process. To take full advantage of low power consumption and improve the efficiency of these models further, the ...
Jiangrong Shen +5 more
semanticscholar +1 more source
GSLB: The Graph Structure Learning Benchmark [PDF]
Neural Information Processing Systems, 2023 Graph Structure Learning (GSL) has recently garnered considerable attention due to its ability to optimize both the parameters of Graph Neural Networks (GNNs) and the computation graph structure simultaneously.
Zhixun Li +10 more
semanticscholar +1 more source
Learning Latent Jet Structure [PDF]
Symmetry, 2021 We summarize our recent work on how to infer on jet formation processes directly from substructure data using generative statistical models. We recount in detail how to cast jet substructure observables’ measurements in terms of Bayesian mixed membership models, in particular Latent Dirichlet Allocation.
Dillon, Barry M +3 more
openaire +3 more sources
Structure-preserving deep learning [PDF]
European Journal of Applied Mathematics, 2021 Over the past few years, deep learning has risen to the foreground as a topic of massive interest, mainly as a result of successes obtained in solving large-scale image processing tasks. There are multiple challenging mathematical problems involved in applying deep learning: most deep learning methods require the solution of hard optimisation problems,
CELLEDONI, E. +6 more
openaire +5 more sources
OpenGSL: A Comprehensive Benchmark for Graph Structure Learning [PDF]
Neural Information Processing Systems, 2023 Graph Neural Networks (GNNs) have emerged as the de facto standard for representation learning on graphs, owing to their ability to effectively integrate graph topology and node attributes.
Zhiyao Zhou +9 more
semanticscholar +1 more source
Adaptive learning is structure learning in time. [PDF]
Neurosci Biobehav Rev, 2021 People use information flexibly. They often combine multiple sources of relevant information over time in order to inform decisions with little or no interference from intervening irrelevant sources. They adjust the degree to which they use new information over time rationally in accordance with environmental statistics and their own uncertainty.
Yu LQ, Wilson RC, Nassar MR.
europepmc +4 more sources
Structure Learning with Continuous Optimization: A Sober Look and Beyond [PDF]
CLEaR, 2023 This paper investigates in which cases continuous optimization for directed acyclic graph (DAG) structure learning can and cannot perform well and why this happens, and suggests possible directions to make the search procedure more reliable.
Ignavier Ng, Biwei Huang, Kun Zhang
semanticscholar +1 more source
Graph Structure Learning for Robust Graph Neural Networks [PDF]
Knowledge Discovery and Data Mining, 2020 Graph Neural Networks (GNNs) are powerful tools in representation learning for graphs. However, recent studies show that GNNs are vulnerable to carefully-crafted perturbations, called adversarial attacks.
Wei Jin +5 more
semanticscholar +1 more source