Search results for "NEURAL NETWORK"
showing 10 items of 1385 documents
A Deep Reinforcement Learning scheme for Battery Energy Management
2020
Deep reinforcement learning is considered promising for many energy cost optimization tasks in smart buildings. How-ever, agent learning, in this context, is sometimes unstable and unpredictable, especially when the environments are complex. In this paper, we examine deep Reinforcement Learning (RL) algorithms developed for game play applied to a battery control task with an energy cost optimization objective. We explore how agent behavior and hyperparameters can be analyzed in a simplified environment with the goal of modifying the algorithms for increased stability. Our modified Deep Deterministic Policy Gradient (DDPG) agent is able to perform consistently close to the optimum over multi…
On Analytical vs . Schizophrenic Procedures for Computing Music
2009
The authors present a perspective on computer music, which is based on some particular definitions of music in relation to oral culture and cybernetics. They describe some experiments with different models of neural architectures which generate original music, and then suggest that if such neural systems are rich, effective and intuitive enough to produce ‘live’ music, the understanding of their behaviour may require the development of some ‘schizophrenic’ procedures, as well as analytical ones.
Molecular mode-coupling theory applied to a liquid of diatomic molecules
2000
We study the molecular mode coupling theory for a liquid of diatomic molecules. The equations for the critical tensorial nonergodicity parameters ${\bf F}_{ll'}^m(q)$ and the critical amplitudes of the $\beta$ - relaxation ${\bf H}_{ll'}^m(q)$ are solved up to a cut off $l_{co}$ = 2 without any further approximations. Here $l,m$ are indices of spherical harmonics. Contrary to previous studies, where additional approximations were applied, we find in agreement with simulations, that all molecular degrees of freedom vitrify at a single temperature $T_c$. The theoretical results for the non ergodicity parameters and the critical amplitudes are compared with those from simulations. The qualitat…
Segmentation-Free Estimation of Aortic Diameters from MRI Using Deep Learning
2021
Accurate and reproducible measurements of the aortic diameters are crucial for the diagnosis of cardiovascular diseases and for therapeutic decision making. Currently, these measurements are manually performed by healthcare professionals, being time consuming, highly variable, and suffering from lack of reproducibility. In this work we propose a supervised deep-learning method for the direct estimation of aortic diameters. The approach is devised and tested over 100 magnetic resonance angiography scans without contrast agent. All data was expert-annotated at six aortic locations typically used in clinical practice. Our approach makes use of a 3D+2D convolutional neural network (CNN) that ta…
An implicitly parallel EDA based on restricted boltzmann machines
2014
We present a parallel version of RBM-EDA. RBM-EDA is an Estimation of Distribution Algorithm (EDA) that models dependencies between decision variables using a Restricted Boltzmann Machine (RBM). In contrast to other EDAs, RBM-EDA mainly uses matrix-matrix multiplications for model estimation and sampling. Hence, for implementation, standard libraries for linear algebra can be used. This allows an easy parallelization and leads to a high utilization of parallel architectures. The probabilistic model of the parallel version and the version on a single core are identical. We explore the speedups gained from running RBM-EDA on a Graphics Processing Unit. For problems of bounded difficulty like …
On the Robustness of Deep Features for Audio Event Classification in Adverse Environments
2018
Deep features, responses to complex input patterns learned within deep neural networks, have recently shown great performance in image recognition tasks, motivating their use for audio analysis tasks as well. These features provide multiple levels of abstraction which permit to select a sufficiently generalized layer to identify classes not seen during training. The generalization capability of such features is very useful due to the lack of complete labeled audio datasets. However, as opposed to classical hand-crafted features such as Mel-frequency cepstral coefficients (MFCCs), the performance impact of having an acoustically adverse environment has not been evaluated in detail. In this p…
An integrated framework for risk profiling of breast cancer patients following surgery.
2006
Objective: An integrated decision support framework is proposed for clinical oncologists making prognostic assessments of patients with operable breast cancer. The framework may be delivered over a web interface. It comprises a triangulation of prognostic modelling, visualisation of historical patient data and an explanatory facility to interpret risk group assignments using empirically derived Boolean rules expressed directly in clinical terms. Methods and materials: The prognostic inferences in the interface are validated in a multicentre longitudinal cohort study by modelling retrospective data from 917 patients recruited at Christie Hospital, Wilmslow between 1983 and 1989 and predictin…
A Neural Architecture for Segmentation and Modelling of Range Data
2003
A novel, two stage, neural architecture for the segmentation of range data and their modeling with undeformed superquadrics is presented. The system is composed by two distinct neural stages: a SOM is used to perform data segmentation, and, for each segment, a multi-layer feed-forward network performs model estimation. The topology preserving nature of the SOM algorithm makes this architecture suited to cluster data with respect to sudden curvature variations. The second stage is designed to model and compute the inside-outside function of an undeformed superquadric in whatever attitude, starting form the (x, y, z) data triples. The network has been trained using backpropagation, and the we…
Modelling net radiation at surface using “in situ” netpyrradiometer measurements with artificial neural networks
2011
The knowledge of net radiation at the surface is of fundamental importance because it defines the total amount of energy available for the physical and biological processes such as evapotranspiration, air and soil warming. It is measured with net radiometers, but, the radiometers are expensive sensors, difficult to handle, that require constant care and also involve periodic calibration. This paper presents a methodology based on neural networks in order to replace the use of net radiometers (expensive tools) by modeling the relationships between the net radiation and meteorological variables measured in meteorological stations. Two different data sets (acquired at different locations) have…
The relaxation dynamics of a simple glass former confined in a pore
2000
We use molecular dynamics computer simulations to investigate the relaxation dynamics of a binary Lennard-Jones liquid confined in a narrow pore. We find that the average dynamics is strongly influenced by the confinement in that time correlation functions are much more stretched than in the bulk. By investigating the dynamics of the particles as a function of their distance from the wall, we can show that this stretching is due to a strong dependence of the relaxation time on this distance, i.e. that the dynamics is spatially very heterogeneous. In particular we find that the typical relaxation time of the particles close to the wall is orders of magnitude larger than the one of particles …