Real-Time Likelihood-Free Inference of Roman Binary Microlensing Events with Amortized Neural Posterior Estimation.
Keming ZhangJoshua S. BloomB. Scott GaudiFrancois LanusseCasey LamJessica LuPublished in: CoRR (2021)
Keyphrases
- real time
- posterior probability
- neural network
- markov chain monte carlo
- worst case
- maximum likelihood
- video mosaicing
- bayesian networks
- surveillance videos
- probabilistic inference
- intelligent surveillance
- biologically inspired
- parameter estimation
- low cost
- control system
- bayesian framework
- search tree
- posterior distribution
- maximum likelihood estimation
- upper bound
- bayesian model
- importance sampling
- running times
- continuous valued
- event processing
- lower bound