Search results for "Reinforcement"
showing 10 items of 230 documents
Reinforcement Learning Based Mobility Load Balancing with the Cell Individual Offset
2021
In this study, we focus on the cell individual offset (CIO) parameter in the handover process, which represents the willingness of a cell to admit the incoming handovers. However, it is challenging to tune the CIO parameter, as any poor implementation can lead to undesired outcomes, such as making the neighboring cells over-loaded while decreasing the traffic load of the cell. In this work, a reinforcement learning-based approach for parameter selection is introduced, since it is quite convenient for dynamically changing environments. In that regard, two different techniques, namely Q-learning and SARSA, are proposed, as they are known for their multi-objective optimization capabilities. Mo…
The Dreaming Variational Autoencoder for Reinforcement Learning Environments
2018
Reinforcement learning has shown great potential in generalizing over raw sensory data using only a single neural network for value optimization. There are several challenges in the current state-of-the-art reinforcement learning algorithms that prevent them from converging towards the global optima. It is likely that the solution to these problems lies in short- and long-term planning, exploration and memory management for reinforcement learning algorithms. Games are often used to benchmark reinforcement learning algorithms as they provide a flexible, reproducible, and easy to control environment. Regardless, few games feature a state-space where results in exploration, memory, and plannin…
Multiscale microstructural characterization of particulate-reinforced composite with non-destructive X-ray micro- and nanotomography
2018
Abstract Methods based on X-ray tomography are developed to study the relevant statistical quantities describing the microstructural inhomogeneity of particulate reinforced composites. The developed methods are applied in estimating microstructural inhomogeneity parameters of composites containing metallic glass particles in metal matrix, extruded in varying pressure loads. This study indicates that the critical characteristics with regard to the effect of particle clustering are cluster size and shape, local volume fraction of particles in the cluster and the distance between clusters. The results demonstrate that the spatial distribution of reinforcement is very uneven and the amount of p…
Cross-reinstatement by cocaine and amphetamine of morphine-induced place preference in mice
2005
The cross-reinstatement by psychostimulants of a conditioned place preference (CPP) induced by morphine was evaluated in mice. In Experiment 1, we examined the effects of a single dose of cocaine and amphetamine on a previously extinguished morphine CPP. After acquisition of CPP induced by morphine (40 mg/kg), animals underwent daily extinction sessions of 15 min duration until the CPP was extinguished. Subsequently, animals received a non-contingent injection of cocaine (25 mg/kg) or amphetamine (4 mg/kg), which produced the reinstatement of the extinguished morphine-induced CPP. In Experiment 2, we evaluated the reinstating effects of several priming doses of cocaine (Experiment 2A) or am…
A reinforcement learning approach for individualizing erythropoietin dosages in hemodialysis patients
2009
This paper presents a reinforcement learning (RL) approach for anemia management in patients undergoing chronic renal failure. Erythropoietin (EPO) is the treatment of choice for this kind of anemia but it is an expensive drug and with some dangerous side-effects that should be considered especially for patients who do not respond to the treatment. Therefore, an individualized treatment appears to be necessary. RL is a suitable approach to tackle this problem. Moreover, resulting policies are similar to medical protocols, and hence, they can easily be transferred to daily practice. A cohort of 64 patients are included in the study. An implementation of the Q-learning algorithm based on a st…
Learning to Approach a Moving Ball with a Simulated Two-Wheeled Robot
2006
We show how a two-wheeled robot can learn to approach a moving ball using Reinforcement Learning. The robot is controlled by setting the velocities of its two wheels. It has to reach the ball under certain conditions to be able to kick it towards a given target. In order to kick, the ball has to be in front of the robot. The robot also has to reach the ball at a certain angle in relation to the target, because the ball is always kicked in the direction from the center of the robot to the ball. The robot learns which velocity differences should be applied to the wheels: one of the wheels is set to the maximum velocity, the other one according to this difference. We apply a REINFORCE algorith…
Saying-Doing Correspondence
2002
The study of the correspondence concerns the functional relationships between an individual’s verbal and non-verbal behavior. The analysis of the functional relations between saying and doing is interesting from a theoretical perspective (e.g.: how and when do they relate together? Learning to tell the truth etc.) and from an applied point of view: many clinical procedures, as verbal forms of psychotherapy, are based on the idea that changing people’s verbalizations about their behavior will lead to corresponding changes in the way they behave. Since say-do correspondence training has been employed in a variety of behaviors and types of procedures to examine the conditions upon which the ar…
Nonlinear FE analysis of out-of-plane behaviour of masonry walls with and without CFRP reinforcement
2014
Abstract The out-of-plane behaviour of unreinforced and CFRP reinforced masonry wall is studied by means of experimental investigation and numerical FE modelling. The latter is based on a linear constitutive law both for ashlars and mortar joints constituting masonry and the lines of potential delamination are taken into account by means of an interface element with bi-linear law, reproducing the opening failure mode. When reinforcement is introduced, an interface element with bi-linear law is also used, reproducing sliding failure mode. Comparison between numerical and experimental results show the reliability of the modelling. Moreover, a parametric analysis is carried out in order to inv…
Designing the internal reinforcements of a sailing boat using a topology optimization approach
2022
In naval design it is common practice to define an internal regular web frame made of longitudinal elements and transversal sections with the purpose of giving stiffness to the whole structure and, at the same time, promoting lightness. In this work, FEM simulation and Topology Optimization (TO) tools are implemented to present a different approach in placing the reinforcements inside the hull of a sailing dinghy. The methodology proposed in this paper considers as a starting point the volume inside the hull and the deck completely filled with material and the result after the simulations is a free form shape of the sailboat reinforcements. The TO procedure is based on two different input F…
Virtual Resource Allocation for Wireless Virtualized Heterogeneous Network with Hybrid Energy Supply
2022
In this work, two novel virtual user association and resource allocation algorithms are introduced for a wireless virtualized heterogeneous network with hybrid energy supply. In the considered system, macro base stations (MBSs) are supplied by the grid power and small base stations (SBSs) have the energy harvesting capability in addition to the grid power supplement. Multiple infrastructure providers (InPs) own the physical resources, i.e., BSs and radio resources. The Mobile Virtual Network Operators (MVNOs) are able to recent these resources from the InPs and operate the virtualized resources for providing services to different users. In particular, aiming to maximize the overall utility …