Pliable Rejection Sampling

Akram Erraqabi, Michal Valko, Alexandra Carpentier, Odalric Maillard
Proceedings of The 33rd International Conference on Machine Learning, PMLR 48:2121-2129, 2016.

Abstract

Rejection sampling is a technique for sampling from difficult distributions. However, its use is limited due to a high rejection rate. Common adaptive rejection sampling methods either work only for very specific distributions or without performance guarantees. In this paper, we present pliable rejection sampling (PRS), a new approach to rejection sampling, where we learn the sampling proposal using a kernel estimator. Since our method builds on rejection sampling, the samples obtained are with high probability i.i.d. and distributed according to f. Moreover, PRS comes with a guarantee on the number of accepted samples.

Cite this Paper

BibTeX
@InProceedings{pmlr-v48-erraqabi16, title = {Pliable Rejection Sampling}, author = {Erraqabi, Akram and Valko, Michal and Carpentier, Alexandra and Maillard, Odalric}, booktitle = {Proceedings of The 33rd International Conference on Machine Learning}, pages = {2121--2129}, year = {2016}, editor = {Balcan, Maria Florina and Weinberger, Kilian Q.}, volume = {48}, series = {Proceedings of Machine Learning Research}, address = {New York, New York, USA}, month = {20--22 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v48/erraqabi16.pdf}, url = {https://proceedings.mlr.press/v48/erraqabi16.html}, abstract = {Rejection sampling is a technique for sampling from difficult distributions. However, its use is limited due to a high rejection rate. Common adaptive rejection sampling methods either work only for very specific distributions or without performance guarantees. In this paper, we present pliable rejection sampling (PRS), a new approach to rejection sampling, where we learn the sampling proposal using a kernel estimator. Since our method builds on rejection sampling, the samples obtained are with high probability i.i.d. and distributed according to f. Moreover, PRS comes with a guarantee on the number of accepted samples.} }
Endnote
%0 Conference Paper %T Pliable Rejection Sampling %A Akram Erraqabi %A Michal Valko %A Alexandra Carpentier %A Odalric Maillard %B Proceedings of The 33rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2016 %E Maria Florina Balcan %E Kilian Q. Weinberger %F pmlr-v48-erraqabi16 %I PMLR %P 2121--2129 %U https://proceedings.mlr.press/v48/erraqabi16.html %V 48 %X Rejection sampling is a technique for sampling from difficult distributions. However, its use is limited due to a high rejection rate. Common adaptive rejection sampling methods either work only for very specific distributions or without performance guarantees. In this paper, we present pliable rejection sampling (PRS), a new approach to rejection sampling, where we learn the sampling proposal using a kernel estimator. Since our method builds on rejection sampling, the samples obtained are with high probability i.i.d. and distributed according to f. Moreover, PRS comes with a guarantee on the number of accepted samples.
RIS
TY - CPAPER TI - Pliable Rejection Sampling AU - Akram Erraqabi AU - Michal Valko AU - Alexandra Carpentier AU - Odalric Maillard BT - Proceedings of The 33rd International Conference on Machine Learning DA - 2016/06/11 ED - Maria Florina Balcan ED - Kilian Q. Weinberger ID - pmlr-v48-erraqabi16 PB - PMLR DP - Proceedings of Machine Learning Research VL - 48 SP - 2121 EP - 2129 L1 - http://proceedings.mlr.press/v48/erraqabi16.pdf UR - https://proceedings.mlr.press/v48/erraqabi16.html AB - Rejection sampling is a technique for sampling from difficult distributions. However, its use is limited due to a high rejection rate. Common adaptive rejection sampling methods either work only for very specific distributions or without performance guarantees. In this paper, we present pliable rejection sampling (PRS), a new approach to rejection sampling, where we learn the sampling proposal using a kernel estimator. Since our method builds on rejection sampling, the samples obtained are with high probability i.i.d. and distributed according to f. Moreover, PRS comes with a guarantee on the number of accepted samples. ER -
APA
Erraqabi, A., Valko, M., Carpentier, A. & Maillard, O.. (2016). Pliable Rejection Sampling. Proceedings of The 33rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 48:2121-2129 Available from https://proceedings.mlr.press/v48/erraqabi16.html.

Related Material