@inproceedings{discovery10056434, pages = {113.1--113.15}, booktitle = {Proceedings of the 45th International Colloquium on Automata, Languages, and Programming (ICALP 2018)}, volume = {107}, address = {Prague, Czech Republic}, note = {{\copyright} Alejandro Aguirre, Gilles Barthe, Justin Hsu, and Alexandra Silva; licensed under Creative Commons License CC-BY}, publisher = {LIPIcs , Schloss Dagstuhl - Leibniz-Zentrum f{\"u}r Informatik}, year = {2018}, title = {Almost sure productivity}, journal = {Leibniz International Proceedings in Informatics, LIPIcs}, month = {July}, keywords = {Coinduction, Probabilistic Programming, Productivity}, url = {https://discovery.ucl.ac.uk/id/eprint/10056434/}, issn = {1868-8969}, author = {Aguirre, A and Barthe, G and Hsu, J and Silva, A}, abstract = {We introduce Almost Sure Productivity (ASP), a probabilistic generalization of the productivity condition for coinductively defined structures. Intuitively, a probabilistic coinductive stream or tree is ASP if it produces infinitely many outputs with probability 1. Formally, we define ASP using a final coalgebra semantics of programs inspired by Kerstan and K{\"o}nig. Then, we introduce a core language for probabilistic streams and trees, and provide two approaches to verify ASP: a syntactic sufficient criterion, and a decision procedure by reduction to model-checking LTL formulas on probabilistic pushdown automata.} }