Gradient Estimation Using Stochastic Computation Graphs

@inproceedings{Schulman2015GradientEU,
  title={Gradient Estimation Using Stochastic Computation Graphs},
  author={John Schulman and Nicolas Heess and Theophane Weber and Pieter Abbeel},
  booktitle={NIPS},
  year={2015}
}
In a variety of problems originating in supervised, unsupervised, and reinforcement learning, the loss function is defined by an expectation over a collection of random variables, which might be part of a probabilistic model or the external world. Estimating the gradient of this loss function, using samples, lies at the core of gradient-based learning algorithms for these problems. We introduce the formalism of stochastic computation graphs—directed acyclic graphs that include both… CONTINUE READING
Highly Cited
This paper has 129 citations. REVIEW CITATIONS

Citations

Publications citing this paper.
Showing 1-10 of 94 extracted citations

129 Citations

0204060'14'16'18
Citations per Year
Semantic Scholar estimates that this publication has 129 citations based on the available data.

See our FAQ for additional information.

References

Publications referenced by this paper.
Showing 1-10 of 30 references

Monte Carlo methods in financial engineering

P. Glasserman
volume 53. Springer Science & Business Media • 2003
View 5 Excerpts
Highly Influenced

Auto-Encoding Variational Bayes

View 5 Excerpts
Highly Influenced

Gradient estimation

M. C. Fu
Handbooks in operations research and management science, 13:575–616 • 2006
View 5 Excerpts
Highly Influenced

Similar Papers

Loading similar papers…