Jakob Carlström

Learn More
— This paper presents an architecture and algorithms for optimizing the performance of web services. For a given service , session-based admission control is combined with stage-wise request queuing, where the stages represent sub-tasks within sessions. The scheduling of requests is governed by generalized processor sharing. We present a performance model,(More)
This paper presents predictive gain scheduling, a technique for simplifying reinforcement learning problems by decomposition. Link admission control of self-similar call traffic is used to demonstrate the technique. The control problem is decomposed into on-line prediction of near-future call arrival rates, and precomputation of policies for Poisson call(More)
SUMMARY An optimal solution of the call admission control and routing problem in multi-service loss networks, in terms of average reward per time unit, is possible by modeling the network behavior as a Markov decision process (MDP). However, even after applying the standard link independence assumption, the solution of the corresponding set of link problems(More)
This paper presents an adaptive scheme for a sub-function in Asynchronous Transfer Mode (ATM) network routing, called link allocation. The scheme adapts the link allocation policy to the offered Poisson call traffic such that the long-term revenue in maximized. It decomposes the link allocation task into a set of link admission control (LAC) tasks,(More)
The robots designed by Autonomous relief (RFC Uppsala) constitute a fully autonomous distributed robot system for solving rescue tasks. This year's robots are much improved, with a new powerful AI engine, more sensors and more accurate positioning routines. The new AI incorporates all sensor data available to all robots in a team into one accurate map. Dead(More)
Quantization errors appear in digital implementations of Multi Layer Perceptrons (MLPs), whose weight and signal values are represented as binary integers of limited length. This paper discusses how to select the representation of weights and how to modify the training in a way that minimizes such errors. Experiments on MLPs for sonar data classification(More)
  • 1