Etalumis: bringing probabilistic programming to scientific simulators at scale

@article{Baydin2019EtalumisBP,
  title={Etalumis: bringing probabilistic programming to scientific simulators at scale},
  author={Atilim Gunes Baydin and Lei Shao and W. Bhimji and L. Heinrich and Lawrence Meadows and Jialin Liu and Andreas Munk and Saeid Naderiparizi and Bradley Gram-Hansen and Gilles Louppe and Mingfei Ma and X. Zhao and P. Torr and V. Lee and K. Cranmer and Prabhat and Frank D. Wood},
  journal={Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis},
  year={2019}
}
  • Atilim Gunes Baydin, Lei Shao, +14 authors Frank D. Wood
  • Published 2019
  • Computer Science, Mathematics
  • Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis
Probabilistic programming languages (PPLs) are receiving widespread attention for performing Bayesian inference in complex generative models. However, applications to science remain limited because of the impracticability of rewriting complex scientific simulators in a PPL, the computational cost of inference, and the lack of scalable implementations. To address these, we present a novel PPL framework that couples directly to existing scientific simulators through a cross-platform probabilistic… Expand
25 Citations
Deep Probabilistic Surrogate Networks for Universal Simulator Approximation
  • 5
  • PDF
SBI - A toolkit for simulation-based inference
  • 11
  • PDF
A Probabilistic Machine Learning Approach to Scheduling Parallel Loops With Bayesian Optimization
Approximate Inference in Discrete Distributions with Monte Carlo Tree Search and Value Functions
  • 1
  • PDF
Learning Proposals for Probabilistic Programs with Inference Combinators
  • Highly Influenced
  • PDF
Amortized Rejection Sampling in Universal Probabilistic Programming
  • 5
  • PDF
Simulation-based inference methods for particle physics
  • 3
  • PDF
The Idea of a Probabilistic Programming Language
  • PDF
...
1
2
3
...

References

SHOWING 1-10 OF 11 REFERENCES
Automatic inference for higher-order probabilistic programs
  • 4
  • Highly Influential
Inference Compilation and Universal Probabilistic Programming
  • 97
  • Highly Influential
  • PDF
Edward: A library for probabilistic modeling, inference, and criticism
  • 237
  • Highly Influential
  • PDF
MCMC Using Hamiltonian Dynamics
  • 1,838
  • Highly Influential
  • PDF
Long Short-Term Memory
  • 38,119
  • Highly Influential
  • PDF
Large Batch Training of Convolutional Networks
  • 212
  • Highly Influential
  • PDF
Event generation with SHERPA 1.1
  • 1,996
  • Highly Influential
  • PDF
A Tutorial on Particle Filtering and Smoothing: Fifteen years later
  • 1,708
  • Highly Influential
  • PDF
GradientBased Learning Applied to Document Recognition
  • 2,865
  • Highly Influential
Long short-termmemory
  • Neural computation 9,
  • 1997
...
1
2
...