Scaling up Bayesian Uncertainty Quantification for Inverse Problems Using Deep Neural Networks

Mathematics

• Uncertainty Quantification 86%
• Neural Networks 61%
• Emulation 60%
• Scaling 55%
• Inverse Problem 52%
• Bayesian inference 41%
• Calibration 28%
• Speedup 27%
• High-dimensional 21%
• Surrogate Model 19%
• Nonlinear Inverse Problems 17%
• Linear Inverse Problems 16%
• Advection-diffusion 16%
• Neural Network Model 16%
• Monte Carlo Algorithm 15%
• Dimension Reduction 14%
• Training 13%
• Applied mathematics 13%
• Markov Chain Monte Carlo 13%
• Bayesian Approach 12%
• Physics 10%
• Engineering 10%
• Tend 9%
• Numerical Examples 8%
• Demonstrate 7%

Business & Economics

• Inverse Problem 100%
• Neural Networks 77%
• Scaling 77%
• Quantification 68%
• Bayesian Inference 57%
• Emulation 54%
• Uncertainty 38%
• Sampling 27%
• Calibration 26%
• Dimension Reduction 25%
• Markov Chain Monte Carlo 15%
• Physics 14%
• Bayesian Approach 14%
• Mathematics 13%
• Network Model 12%
• Inference 10%

Engineering & Materials Science

• Inverse problems 70%
• Deep neural networks 63%
• Uncertainty 40%
• Calibration 33%
• Sampling 30%
• Advection 27%
• Markov chains 21%
• Physics 18%