Scalability White Papers

VDCBPI: An Approximate Scalable Algorithm for Large POMDPs

Overview Existing algorithms for discrete partially observable Markov decision processes can at best solve problems of a few thousand states due to two important sources of intractability: the curse of dimensionality and the policy space complexity. This paper describes a new algorithm (VDCBPI) that mitigates both sources of intractability by combining the Value Directed Compression (VDC) technique with Bounded Policy Iteration (BPI). The scalability of VDCBPI is demonstrated on synthetic network management problems with up to 33 million states.

Further White Paper Details
PublisherUniversity of Toronto File FormatPDF
Date PublishedJanuary 2008
FormatWhite Papers   
Topics

Contact Centres: Optimum service at optimum cost

Getting the balance right between meeting the inbound call expectations of busy customers and optimising telecoms costs is the goal for many ICT managers. Technology now enables far more powerful...

NHS EMIS - Intel? Xeon? processors Transform National Healthcare Supplier's Data CentreNHS EMIS - Intel? Xeon? processors Transform National Healthcare Supplier's Data Centre

EMIS is the UK's leading IT supplier in primary healthcare. As the market segment leader for general practice systems, it is widely acknowledged as being at the forefront of medical...

Citizants: Bringing IT In-House

Citizant enhances IT flexibility and scalability by creating a virtualized Dell infrastructure that will pay for itself in 18 months.

Collaborating on Cost Savings: A Scalable Software Whitepaper

It should surely therefore come as no surprise to anyone that IT executives are unlikely to own up to years of unnecessary expenditure, indeed it is more likely that they...

Gobierno del Estado de Veracruz, Mexico Improves Analysis With Integrated Information System

Gobierno del Estado de Veracruz wanted to create a robust, flexible, and scalable information system to facilitate information gathering and provide access to up-to-date information for objective decision making and...


Quick Sitemap Links: