Last edited by Kajirr
Saturday, May 9, 2020 | History

2 edition of Exploiting structure to efficiently solve large scale partially observable Markov decision processes. found in the catalog.

Exploiting structure to efficiently solve large scale partially observable Markov decision processes.

Pascal Poupart

Exploiting structure to efficiently solve large scale partially observable Markov decision processes.

by Pascal Poupart

  • 48 Want to read
  • 9 Currently reading

Published .
Written in English


About the Edition

Partially observable Markov decision processes (POMDPs) provide a natural and principled framework to model a wide range of sequential decision making problems under uncertainty. To date, the use of POMDPs in real-world problems has been limited by the poor scalability of existing solution algorithms, which can only solve problems with up to ten thousand states. In fact, the complexity of finding an optimal policy for a finite-horizon discrete POMDP is PSPACE-complete. In practice, two important sources of intractability plague most solution algorithms: Large policy spaces and large state spaces.In practice, it is critical to simultaneously mitigate the impact of complex policy representations and large state spaces. Hence, this thesis describes three approaches that combine techniques capable of dealing with each source of intractability: VDC with BPI, VDC with Perseus (a randomized point-based value iteration algorithm by Spaan and Vlassis [136]), and state abstraction with Perseus. The scalability of those approaches is demonstrated on two problems with more than 33 million states: synthetic network management and a real-world system designed to assist elderly persons with cognitive deficiencies to carry out simple daily tasks such as hand-washing. This represents an important step towards the deployment of POMDP techniques in ever larger, real-world, sequential decision making problems.On the other hand, for many real-world POMDPs it is possible to define effective policies with simple rules of thumb. This suggests that we may be able to find small policies that are near optimal. This thesis first presents a Bounded Policy Iteration (BPI) algorithm to robustly find a good policy represented by a small finite state controller. Real-world POMDPs also tend to exhibit structural properties that can be exploited to mitigate the effect of large state spaces. To that effect, a value-directed compression (VDC) technique is also presented to reduce POMDP models to lower dimensional representations.

The Physical Object
Pagination144 leaves.
Number of Pages144
ID Numbers
Open LibraryOL19475491M
ISBN 100494027274


Share this book
You might also like
Blast furnace blowing engines

Blast furnace blowing engines

United States maverick tokens

United States maverick tokens

1662 Royal Brevier Prayer Book with Hymns Ancient and Modern Red imitation leather, 4920 2482

1662 Royal Brevier Prayer Book with Hymns Ancient and Modern Red imitation leather, 4920 2482

Alanna

Alanna

Small unit action in Vietnam, summer 1966

Small unit action in Vietnam, summer 1966

Dialects in decline

Dialects in decline

A collection of poetical essays

A collection of poetical essays

Old English mansions

Old English mansions

Discover Reading

Discover Reading

Optimum and limiting hydrogen-ion concentrations for B. botulinus and quantitative estimation of its growth

Optimum and limiting hydrogen-ion concentrations for B. botulinus and quantitative estimation of its growth

Human disease.

Human disease.

Recent theistic discussion

Recent theistic discussion

Dissertations on the subjects and mode of Gospel baptism

Dissertations on the subjects and mode of Gospel baptism

future of the Indian past

future of the Indian past

Instructors manual

Instructors manual

Abraham Lincoln, 1809-1858.

Abraham Lincoln, 1809-1858.

Exploiting structure to efficiently solve large scale partially observable Markov decision processes by Pascal Poupart Download PDF EPUB FB2