The impact of low-cost molecular geometry optimization in property prediction via graph neural network.
Gabriel A. PinheiroFelipe V. CalderanJuarez L. F. Da SilvaMarcos G. QuilesPublished in: ICMLA (2022)
Keyphrases
- low cost
- neural network
- prediction model
- three dimensional
- protein function prediction
- optimization algorithm
- artificial neural networks
- graph properties
- graph representation
- global optimization
- prediction accuracy
- random walk
- fuzzy logic
- image reconstruction from projections
- elman network
- chaotic time series
- topological information
- graph theory
- multi layer perceptron
- neural network ensemble
- prediction error
- prediction algorithm
- drug discovery
- graph model
- pattern recognition
- weighted graph
- back propagation
- graph structure
- structured data
- neural network model
- bp neural network
- social networks
- optimization problems
- highly non linear
- optimization method
- drug design
- protein structure prediction
- graph matching
- highly efficient
- graph theoretic
- predictive model
- bipartite graph
- spanning tree
- graph mining
- constrained optimization
- directed acyclic graph
- network model