Sökning: "Learning Dynamical Models"

Visar resultat 1 - 5 av 22 avhandlingar innehållade orden Learning Dynamical Models.

1. 1. Identification of Stochastic Nonlinear Dynamical Models Using Estimating Functions

   Författare :Mohamed Abdalmoaty; Håkan Hjalmarsson; Adrian Wills; KTH; []
   Nyckelord :TEKNIK OCH TEKNOLOGIER; ENGINEERING AND TECHNOLOGY; Prediction Error Method; Maximum Likelihood; Data-driven; Learning; Stochastic; Nonlinear; Dynamical Models; Non-stationary Linear Predictors; Intractable Likelihood; Latent Variable Models; Estimation; Process Disturbance; Electrical Engineering; Elektro- och systemteknik;

   Sammanfattning : Data-driven modeling of stochastic nonlinear systems is recognized as a very challenging problem, even when reduced to a parameter estimation problem. A main difficulty is the intractability of the likelihood function, which renders favored estimation methods, such as the maximum likelihood method, analytically intractable. LÄS MER

3. 2. Learning from Interactions : Forward and Inverse Decision-Making for Autonomous Dynamical Systems

   Författare :Inês de Miranda de Matos Lourenço; Bo Wahlberg; Sandra Hirche; KTH; []
   Nyckelord :TEKNIK OCH TEKNOLOGIER; ENGINEERING AND TECHNOLOGY; Intelligent systems; autonomous decision-making; Reinforcement Learning; Markov models; Human-Robot Interaction; Biologically-inspired systems; Electrical Engineering; Elektro- och systemteknik;

   Sammanfattning : Decision-making is the mechanism of using available information to generate solutions to given problems by forming preferences, beliefs, and selecting courses of action amongst several alternatives. In this thesis, we study the mechanisms that generate behavior (the forward problem) and how their characteristics can explain observed behavior (the inverse problem). LÄS MER

4. 3. Reinforcement Learning and Dynamical Systems

   Författare :Björn Lindenberg; Karl-Olof Lindahl; Marc G. Bellemare; Linnéuniversitetet; []
   Nyckelord :NATURVETENSKAP; NATURAL SCIENCES; artificial intelligence; distributional reinforcement learning; Markov decision processes; Bellman operators; deep learning; multi-armed bandits; Bayesian bandits; conjugate priors; Thompson sampling; linear finite dynamical systems; cycle orbits; fixed-point systems; Mathematics; Matematik; Computer Science; Datavetenskap;

   Sammanfattning : This thesis concerns reinforcement learning and dynamical systems in finite discrete problem domains. Artificial intelligence studies through reinforcement learning involves developing models and algorithms for scenarios when there is an agent that is interacting with an environment. LÄS MER

5. 4. Bayesian learning of structured dynamical systems

   Författare :Riccardo Sven Risuleo; Håkan Hjalmarsson; Johan Schoukens; KTH; []
   Nyckelord :TEKNIK OCH TEKNOLOGIER; ENGINEERING AND TECHNOLOGY; system identification; bayesian learning; machine learning; Gaussian processes; Electrical Engineering; Elektro- och systemteknik;

   Sammanfattning : In this thesis, we propose some Bayesian approaches to the identificationof structured dynamical systems. In particular, we consider block-orientedmodels in which a complex system is built starting from simple linear andnonlinear building blocks. LÄS MER

7. 5. Learning Stochastic Nonlinear Dynamical Systems Using Non-stationary Linear Predictors

   Författare :Mohamed Abdalmoaty; Håkan Hjalmarsson; Jimmy Olsson; KTH; []
   Nyckelord :TEKNIK OCH TEKNOLOGIER; ENGINEERING AND TECHNOLOGY; TEKNIK OCH TEKNOLOGIER; ENGINEERING AND TECHNOLOGY; TEKNIK OCH TEKNOLOGIER; ENGINEERING AND TECHNOLOGY; NATURVETENSKAP; NATURAL SCIENCES; Stochastic Nonlinear Systems; Nonlinear System Identification; Learning Dynamical Models; Maximum Likelihood; Estimation; Process Disturbance; Prediction Error Method; Non-stationary Linear Predictors; Intractable Likelihood; Latent Variable Models; Electrical Engineering; Elektro- och systemteknik;

   Sammanfattning : The estimation problem of stochastic nonlinear parametric models is recognized to be very challenging due to the intractability of the likelihood function. Recently, several methods have been developed to approximate the maximum likelihood estimator and the optimal mean-square error predictor using Monte Carlo methods. LÄS MER