Master thesis electrical power

Inverse stochastic optimal here, in the context of MDPs, is an active research master thesis and has a wide spectrum of applications in electrical power fields such as engineering, operations research and biology.

There are two main motivations behind inverse decision making.

The first one concerns situations where power cost function is of interest by itself, e. The second one concerns master thesis electrical power task of imitation or apprenticeship learning by first recovering the expert's cost function and then using it to reproduce and synthesize the optimal behavior.

For instance, power engineering, inverse MDPs can be used to explain power imitate the observed expert behavior, e. Power examples master thesis electrical power be found in humanoid robotics and understanding of here locomotion.

Navigation Area

Most of existing inverse MDP methods, either are designed exclusively for MDPs with finite state and action spaces, or rely on an oracle access to an MDP solver which is used in the college essay majors loop of an iterative procedure. This introduces a significant computational burden in applying these methods to MDPs with continuous state and action spaces, since solving a continuous MDP is a challenging and computationally expensive problem on its own, especially in the case of unknown dynamics.

As a result, inverse optimal control for MDPs over uncountable power remains master thesis electrical power unexplored. The goal of this project is to contribute in this line master thesis electrical power power. Under the assumption that the control model is polynomial, the student will propose a method that avoids solving repeatedly the forward problem and at the same master thesis electrical power provides probabilistic performance guarantees on the master thesis electrical of the master thesis electrical solution.

The approach will master thesis electrical power based on the LP formulation of MDPs, complementary slackness optimality conditions, recent developments in polynomial optimization and uniform finite sample master thesis electrical from statistical learning theory.

The relevance of the developed approach will be illustrated on an autonomous flight problem, first in simulation, and if time permits also in experiment, using the exisiting quadrotor test-bed available at IfA.

Goal The power objectives of this master thesis electrical are the following. These include polynomial optimization sum-of-squares programming, linear matrix inequalitiesthe LP approach to MDPs, as well as tools from statistical learning theory; - gain experience in doing research.

In particular the student will develop mathematical proofs and formal logic arguments; power be familiar with the use of semidefinite programming SDP solvers e. The project is well suited power a student who enjoys mathematics.