DeepParticle: learning invariant measure by a deep neural network minimizing Wasserstein distance on data generated from an interacting particle method.
Zhongjian WangJack XinZhiwen ZhangPublished in: CoRR (2021)
Keyphrases
- image registration
- affine transformation
- synthetic data
- neural network
- prior knowledge
- distance matrix
- neural nets
- test data
- similarity measure
- data sets
- correlation analysis
- training data
- learning process
- bayesian methods
- data analysis
- missing values
- statistical methods
- background knowledge
- missing data
- unsupervised learning
- data structure
- learned models
- input data
- feedforward neural networks
- noisy data
- active learning
- learning algorithm
- bp neural network
- pairwise
- feed forward neural networks
- hamming distance
- dissimilarity measure
- distance function
- correlation coefficient
- data sources
- objective function
- distance metric
- reinforcement learning