Adaptive Importance Sampling from Probabilistic Tree Automata

Christoph Teichmann, Kasimir Wansing, and Alexander Koller

In Proceedings of the ACL Workshop on Statistical NLP and Weighted Automata, Berlin, 2016.

We present a general importance sampling technique for approximating expected values based on samples from probabilistic finite tree automata. The algorithm uses the samples it produces to adapt rule probabilities for the automaton in order to improve sample quality.

Download: Download

BibTeX Entry
@InProceedings{sgd-ais-16,
	Title = {Adaptive Importance Sampling from Probabilistic Tree Automata},
	Author = {Christoph Teichmann and Kasimir Wansing and Alexander Koller},
	Booktitle = {Proceedings of the ACL Workshop on Statistical NLP and Weighted Automata},
	Year = {2016},
	address = {Berlin}
}

Back: Publications