Links between multiplicity automata, observable operator models and predictive state representations -- a unified learning framework

MaRDI QIDQ5502106zbMATH OpenFDO

Authors Michael Thon, Herbert Jaeger

Publication date 17 August 2015

Full work available at URL http://jmlr.csail.mit.edu/papers/v16/thon15a.html

zbMATH Keywords

hidden Markov modelsmultiplicity automataobservable operator modelspredictive state representationsspectral learning algorithms

Mathematics Subject Classification ID

Computational learning theory (68Q32) Formal languages and automata (68Q45) Algebraic theory of languages and automata (68Q70) Learning and adaptive systems in artificial intelligence (68T05)

Recommendations

• Erratum (“Fast and Robust Learning by Reinforcement Signals: Explorations in the Insect Brain” by Ramón Huerta and Thomas Nowotny, Neural Computation, August 2009, Vol. 21, No. 8: 2123–2151)
• Learning Theory
• Norm-observable operator models
• Links between probabilistic automata and hidden Markov models: probability distributions, learning models and induction algorithms
• Learning probability distributions generated by finite-state machines

Cited in

(4)

• Topology, convergence, and reconstruction of predictive states
• Norm-observable operator models
• Making the error-controlling algorithm of observable operator models constructive
• Exploring predictive states via Cantor embeddings and Wasserstein distance

Describes a project that uses

Uses Software

• Regularization tools
• UTV

This page was built for publication: Links between multiplicity automata, observable operator models and predictive state representations -- a unified learning framework

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5502106)