Identifiability of Model Properties in Over-Parameterized Model Classes

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

Abstract

Classical learning theory is based on a tight linkage between hypothesis space (a class of function on a domain X), data space
(function-value examples (x, f(x))), and the space of queries for the learned model (predicting function values for new examples x). However, in many learning scenarios the 3-way association between hypotheses, data, and queries can really be much looser. Model classes can be overparameterized, i.e., different hypotheses may be equivalent with respect to the data observations. Queries may relate to model properties that do not directly correspond to the observations in the data. In this paper we make some initial steps to extend and adapt basic concepts of computational learnability and statistical identifiability to provide a foundation for investigating learnability in such broader contexts. We exemplify the use of the framework in three different applications: the identification of temporal logic properties of probabilistic automata learned from sequence data, the identification of causal dependencies in probabilistic graphical models, and the transfer of probabilistic relational models to new domains.
Original languageEnglish
Title of host publicationMachine Learning and Knowledge Discovery in Databases : European Conference, ECML PKDD 2013, Prague, Czech Republic, September 23-27, 2013, Proceedings, Part III
EditorsHendrik Blockeel, Kristian Kersting, Siegfried Nijssen, Filip Železný
Number of pages16
Place of PublicationBerlin
PublisherSpringer Publishing Company
Publication date2013
Pages112-127
ISBN (Print)978-3-642-40993-6
ISBN (Electronic)978-3-642-40994-3
DOIs
Publication statusPublished - 2013
EventECML PKDD 2013 - Prag, Czech Republic
Duration: 23 Sep 201327 Sep 2013

Conference

ConferenceECML PKDD 2013
CountryCzech Republic
CityPrag
Period23/09/201327/09/2013
SeriesLecture Notes in Computer Science
Volume8190
ISSN0302-9743

    Fingerprint

Cite this

Jaeger, M. (2013). Identifiability of Model Properties in Over-Parameterized Model Classes. In H. Blockeel, K. Kersting, S. Nijssen, & F. Železný (Eds.), Machine Learning and Knowledge Discovery in Databases: European Conference, ECML PKDD 2013, Prague, Czech Republic, September 23-27, 2013, Proceedings, Part III (pp. 112-127). Springer Publishing Company. Lecture Notes in Computer Science, Vol.. 8190 https://doi.org/10.1007/978-3-642-40994-3_8