Open Problem: Meeting Times for Learning Random Automata

Benjamin Fish, Lev Reyzin
Proceedings of the 2017 Conference on Learning Theory, PMLR 65:8-11, 2017.

Abstract

Learning automata is a foundational problem in computational learning theory. However, even efficiently learning random DFAs is hard. A natural restriction of this problem is to consider learning random DFAs under the uniform distribution. To date, this problem has no non-trivial lower bounds nor algorithms faster than brute force. In this note, we propose a method to find faster algorithms for this problem. We reduce the learning problem to a conjecture about meeting times of random walks over random DFAs, which may be of independent interest to prove.

Cite this Paper


BibTeX
@InProceedings{pmlr-v65-fish17a, title = {Open Problem: Meeting Times for Learning Random Automata}, author = {Fish, Benjamin and Reyzin, Lev}, booktitle = {Proceedings of the 2017 Conference on Learning Theory}, pages = {8--11}, year = {2017}, editor = {Kale, Satyen and Shamir, Ohad}, volume = {65}, series = {Proceedings of Machine Learning Research}, month = {07--10 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v65/fish17a/fish17a.pdf}, url = {https://proceedings.mlr.press/v65/fish17a.html}, abstract = {Learning automata is a foundational problem in computational learning theory. However, even efficiently learning random DFAs is hard. A natural restriction of this problem is to consider learning random DFAs under the uniform distribution. To date, this problem has no non-trivial lower bounds nor algorithms faster than brute force. In this note, we propose a method to find faster algorithms for this problem. We reduce the learning problem to a conjecture about meeting times of random walks over random DFAs, which may be of independent interest to prove.} }
Endnote
%0 Conference Paper %T Open Problem: Meeting Times for Learning Random Automata %A Benjamin Fish %A Lev Reyzin %B Proceedings of the 2017 Conference on Learning Theory %C Proceedings of Machine Learning Research %D 2017 %E Satyen Kale %E Ohad Shamir %F pmlr-v65-fish17a %I PMLR %P 8--11 %U https://proceedings.mlr.press/v65/fish17a.html %V 65 %X Learning automata is a foundational problem in computational learning theory. However, even efficiently learning random DFAs is hard. A natural restriction of this problem is to consider learning random DFAs under the uniform distribution. To date, this problem has no non-trivial lower bounds nor algorithms faster than brute force. In this note, we propose a method to find faster algorithms for this problem. We reduce the learning problem to a conjecture about meeting times of random walks over random DFAs, which may be of independent interest to prove.
APA
Fish, B. & Reyzin, L.. (2017). Open Problem: Meeting Times for Learning Random Automata. Proceedings of the 2017 Conference on Learning Theory, in Proceedings of Machine Learning Research 65:8-11 Available from https://proceedings.mlr.press/v65/fish17a.html.

Related Material