Recurrence Enhances the Spatial Encoding of Static Inputs in Reservoir Networks
نویسندگان
چکیده
We shed light on the key ingredients of reservoir computing and analyze the contribution of the network dynamics to the spatial encoding of inputs. Therefore, we introduce attractor-based reservoir networks for processing of static patterns and compare their performance and encoding capabilities with a related feedforward approach. We show that the network dynamics improve the nonlinear encoding of inputs in the reservoir state which can increase the task-specific performance.
منابع مشابه
H&infin Controller Design for Static VAR Compensators in Industrial Networks
This paper deals with the design and evaluation of a robust controller for static VAR compensator (SVC) in remote industrial power systems to enhance the voltage profile for three-phase single cage induction motor (SCIM) loads. The controller design is based on H∞ theory to deal with uncertainties arising in industrial network modelling. The performance of the H∞ controller has been evaluated...
متن کاملRock typing and reservoir zonation based on the NMR logging and geological attributes in the mixed carbonate-siliciclastic Asmari Reservoir
Rock typing is known as the best way in heterogeneous reservoirs characterization. The rock typing methods confine to various aspects of the rocks such as multi-scale and multi-modal pore types and size, rock texture, diagenetic modifications and integration of static/dynamic data. Integration of static and dynamic behavior of rocks and their sedimentary features are practiced in this study. Po...
متن کامل3D and 4D Seismic Data Integration in Static and Dynamic Reservoir Modeling: A Review
Reservoir modeling is the process of generating numerical representations of reservoir conditions and properties on the basis of geological, geophysical, and engineering data measured on the Earth’s surface or in depth at a limited number of borehole locations. Therefore, reservoir modeling requires an incorporation of the data from a variety of sources, along with an integration of knowledge a...
متن کاملA Fully Integrated Method for Dynamic Rock Type Characterization Development in One of Iranian Off-Shore Oil Reservoir
Rock selection in modeling and simulation studies is usually based on two techniques; routinely defined rock types and those defined by special core analysis (SCAL). The challenge in utilizing these two techniques is that they are frequently assumed to be the same, but in practice, static rock-types (routinely defined) are not always representative of dynamic rock-types (SCAL defined) in the re...
متن کاملReceptive Field Encoding Model for Dynamic Natural Vision
Introduction: Encoding models are used to predict human brain activity in response to sensory stimuli. The purpose of these models is to explain how sensory information represent in the brain. Convolutional neural networks trained by images are capable of encoding magnetic resonance imaging data of humans viewing natural images. Considering the hemodynamic response function, these networks are ...
متن کامل