It is common practice to use linear plant models and linear controllers in the control systems design. Such approach has simple explanation applying to plants with insignificant nonlinearity or to those, functioning closely to a working point. But linear controllers, indeed with some modification, are used even for plants with significant non-linearities. Because of several reasons the non-line...

We propose a modular reinforcement learning architecture for nonlinear, nonstationary control tasks, which we call multiple model-based reinforcement learning (MMRL). The basic idea is to decompose a complex task into multiple domains in space and time based on the predictability of the environmental dynamics. The system is composed of multiple modules, each of which consists of a state predict...

This paper presents an efficient method for extracting a multi-model interpolation function from a nonlinear system. The multi-model interpolation function consists of couple simplified time-varying models in neural-network structure to dynamically approximate the nature of the physical phenomena to be interpolated and extrapolated. The purpose of using the multi-model interpolation function is...

We introduce the Multiple Quantile Graphical Model (MQGM), which extends the neighborhood selection approach of Meinshausen and Bühlmann for learning sparse graphical models. The latter is defined by the basic subproblem of modeling the conditional mean of one variable as a sparse function of all others. Our approach models a set of conditional quantiles of one variable as a sparse function of ...

Point patterns are sets or multi-sets of unordered points that arise in numerous data analysis problems. This article proposes a framework for model-based point pattern learning using point process theory. Likelihood functions for point pattern data derived from point process theory enable principled yet conceptually transparent extensions of learning tasks, such as classification, novelty dete...

In bricklayers’ model, which is a generalization of the misanthrope processes, we show that a nontrivial class of product distributions is closed under the time-evolution of the process. This class also includes measures fitting to shock data of the limiting PDE. In particular, we show that shocks of this type with discontinuity of size one perform ordinary nearest neighbor random walks only in...

lbe card game Elcllsis [1.51 poscs an induct:on problem in which no single model-bascd gcncralization Sp.1CC capturcs all of the desircd gcncf:lliz.1tions of the input data. A progr:lr:l has becn dcvcloped which solves this problem by se.1(":hing scveral ditTerem genNali7..1tion spaces for pl.111Siblc Elcusis rules. Multiple generalization spaces require multiple data transformation mcthods "fo...

This thesis considers the control of multiple model systems. These are systems for which only one model out of some finite set of models gives the system dynamics at any given time. In particular, the model that gives the system dynamics can change over time. This thesis covers some of the theoretical aspects of these systems, including controllability and stabilizability. As an application, “o...