Bayesian Reinforcement Learning

in Wiering, Marco; van Otterlo, Martijn (Eds.) Reinforcement Learning: State of the Art (2012)

This chapter surveys recent lines of work that use Bayesian techniques for reinforcement learning. In Bayesian learning, uncertainty is expressed by a prior distribution over unknown parameters and ... [more ▼]

This chapter surveys recent lines of work that use Bayesian techniques for reinforcement learning. In Bayesian learning, uncertainty is expressed by a prior distribution over unknown parameters and learning is achieved by computing a posterior distribution based on the data observed. Hence, Bayesian reinforcement learning distinguishes itself from other forms of reinforcement learning by explic- itly maintaining a distribution over various quantities such as the parameters of the model, the value function, the policy or its gradient. This yields several benefits: a) domain knowledge can be naturally encoded in the prior distribution to speed up learning; b) the exploration/exploitation tradeoff can be naturally optimized; and c) notions of risk can be naturally taken into account to obtain robust policies. [less ▲]

Subthalamic nucleus stimulation restores the efferent cortical drive to muscle in parallel to functional motor improvement.

in The European journal of neuroscience (2012), 35(6), 896-908

Pathological synchronization in large-scale motor networks constitutes a pathophysiological hallmark of Parkinson's disease (PD). Corticomuscular synchronization in PD is pronounced in lower frequency ... [more ▼]

Pathological synchronization in large-scale motor networks constitutes a pathophysiological hallmark of Parkinson's disease (PD). Corticomuscular synchronization in PD is pronounced in lower frequency bands (< 10 Hz), whereas efficient cortical motor integration in healthy persons is driven in the beta frequency range. Electroencephalogram and electromyogram recordings at rest and during an isometric precision grip task were performed in four perioperative sessions in 10 patients with PD undergoing subthalamic nucleus deep-brain stimulation: (i) 1 day before (D0); (ii) 1 day after (D1); (iii) 8 days after implantation of macroelectrodes with stimulation off (D8StimOff); and (iv) on (D8StimOn). Analyses of coherence and phase delays were performed in order to challenge the effects of microlesion and stimulation on corticomuscular coherence (CMC). Additionally, local field potentials recorded from the subthalamic nucleus on D1 allowed comprehensive mapping of motor-related synchronization in subthalamocortical and cerebromuscular networks. Motor performance improved at D8StimOn compared with D0 and D8StimOff paralleled by a reduction of muscular activity and CMC in the theta band (3.9-7.8 Hz) and by an increase of CMC in the low-beta band (13.7-19.5 Hz). Efferent motor cortical drives to muscle presented mainly below 10 Hz on D8StimOff that were suppressed on D8StimOn and occurred on higher frequencies from 13 to 45 Hz. On D1, coherence of the high-beta band (20.5-30.2 Hz) increased during movement compared with rest in subthalamomuscular and corticomuscular projections, whereas it was attenuated in subcorticocortical projections. The present findings lend further support to the concept of pathological network synchronization in PD that is beneficially modulated by stimulation. [less ▲]

The 2007‐2009 Financial Crisis: Changing Market Dynamics and the Impact of Credit Supply and Aggregate Demand Sensitivity*

E-print/Working paper (2012)

Improving teachers’ judgments: Accountability affects teachers’ tracking decisions

in International Journal of Technology and Inclusive Education (2012), 1

Emergence of the silicon human and network targeting drugs

in European Journal of Pharmaceutical Sciences (2012), 46(4), 190-197

The development of disease may be characterized as a pathological shift of homeostasis; the main goal of contemporary drug treatment is, therefore, to return the pathological homeostasis back to the ... [more ▼]

The development of disease may be characterized as a pathological shift of homeostasis; the main goal of contemporary drug treatment is, therefore, to return the pathological homeostasis back to the normal physiological range. From the view point of systems biology, homeostasis emerges from the interactions within the network of biomolecules (e.g. DNA, mRNA, proteins), and, hence, understanding how drugs impact upon the entire network should improve their efficacy at returning the network (body) to physiological homeostasis. Large, mechanism-based computer models, such as the anticipated human whole body models (silicon or virtual human), may help in the development of such network-targeting drugs. Using the philosophical concept of weak and strong emergence, we shall here take a more general look at the paradigm of network-targeting drugs, and propose our approaches to scale the strength of strong emergence. We apply these approaches to several biological examples and demonstrate their utility to reveal principles of bio-modeling. We discuss this in the perspective of building the silicon human. [less ▲]

Stochastic POMDP controllers: How easy to optimize?

Scientific Conference (2012)

It was recently shown that computing an optimal stochastic controller in a discounted in finite-horizon partially observable Markov decision process is an NP-hard problem. The reduction (from the ... [more ▼]

It was recently shown that computing an optimal stochastic controller in a discounted in finite-horizon partially observable Markov decision process is an NP-hard problem. The reduction (from the independent-set problem) involves designing an MDP with special state-action rewards. In this note, we show that the case of state-only-dependent rewards is also NP-hard. [less ▲]

Über die Verschriftlichung der Erinnerung in dem Roman La mémoire de la baleine von Jean Portante

in Grucza, Franciszek; Neuberg, Simon; Conter, Claude D. (Eds.) et al Jiddische Spache und Literatur / Luxemburgistik / Niederländische Sprach- und Literaturwissenschaft mit besonderer Berücksichtigung des Afrikaans (2012)

Épreuves Standardisées. Nationaler Bericht 2011‐2012

Book published by University of Luxembourg, EMACS (2012)

War Children in the Post-war. A West-East perspective on child policies, child experiences and warchildhood remembrance cultures in Europe since 1945

Presentation (2012)

East-West Comparisons

Presentation (2012)