Title :

Spectral Learning from a Single Trajectory under Finite-State Policies

Author(s) :

Balle, Borja [Auteur]
Computing Department [Lancaster]
Maillard, Odalric-Ambrym [Auteur]
Sequential Learning [SEQUEL]

Conference title :

International conference on Machine Learning

City :

Sidney

Country :

France

Start date of the conference :

2017-07

Journal title :

Proceedings of the International conference on Machine Learning

Publication date :

2017-07

HAL domain(s) :

Mathématiques [math]/Systèmes dynamiques [math.DS]
Mathématiques [math]/Théorie spectrale [math.SP]
Statistiques [stat]/Machine Learning [stat.ML]

English abstract : [en]

We present spectral methods of moments for learning sequential models from a single trajec-tory, in stark contrast with the classical literature that assumes the availability of multiple i.i.d. trajectories. Our approach ...
Show more >We present spectral methods of moments for learning sequential models from a single trajec-tory, in stark contrast with the classical literature that assumes the availability of multiple i.i.d. trajectories. Our approach leverages an efficient SVD-based learning algorithm for weighted au-tomata and provides the first rigorous analysis for learning many important models using dependent data. We state and analyze the algorithm under three increasingly difficult scenarios: proba-bilistic automata, stochastic weighted automata, and reactive predictive state representations controlled by a finite-state policy. Our proofs include novel tools for studying mixing properties of stochastic weighted automata.Show less >

Language :

Anglais

Peer reviewed article :

Oui

Audience :

Internationale

Popular science :

Non

ANR Project :

BANDITS MANCHOTS POUR SIGNAUX NON-STATIONNAIRES ET STRUCTURES

Collections :

• Centre de Recherche en Informatique, Signal et Automatique de Lille (CRIStAL) - UMR 9189

Source :

Harvested from HAL