Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userApplication Usability Levels: A Framework for Tracking Project Product Progress
84
0
0.0
(
0
)
Ask ChatGPT about the research
No Arabic abstract
The space physics community continues to grow and become both more interdisciplinary and more intertwined with commercial and government operations. This has created a need for a framework to easily identify what projects can be used for specific applications and how close the tool is to routine autonomous or on-demand implementation and operation. We propose the Application Usability Level (AUL) framework and publicizing AULs to help the community quantify the progress of successful applications, metrics, and validation efforts. This framework will also aid the scientific community by supplying the type of information needed to build off of previously published work and publicizing the applications and requirements needed by the user communities. In this paper, we define the AUL framework, outline the milestones required for progression to higher AULs, and provide example projects utilizing the AUL framework. This work has been completed as part of the activities of the Assessment of Understanding and Quantifying Progress working group which is part of the International Forum for Space Weather Capabilities Assessment.
rate research
Read More
CHARA/SPICA (Stellar Parameters and Images with a Cophased Array) is currently being developed at Observatoire de la C^ote dAzur. It will be installed at the visible focus of the CHARA Array by the end of 2021. It has been designed to perform a large survey of fundamental stellar parameters with, in the possible cases, a detailed imaging of the surface or environment of stars. To reach the required precision and sensitivity, CHARA/SPICA combines a low spectral resolution mode R = 140 in the visible and single-mode fibers fed by the AO stages of CHARA. This setup generates additional needs before the interferometric combination: the compensation of atmospheric refraction and longitudinal dispersion, and the fringe stabilization. In this paper, we present the main features of the 6-telescopes fibered visible beam combiner (SPICA-VIS) together with the first laboratory and on-sky results of the fringe tracker (SPICA-FT). We describe also the new fringe-tracker simulator developed in parallel to SPICA-FT.
We present a new method to automatically track filaments over the solar disk. The filaments are first detected on Meudon Spectroheliograph H{alpha} images of the Sun, applying the technique developed by Fuller, Aboudarham, and Bentley (Solar phys. 227, 61, 2005). This technique combines cleaning processes, image segmentation based on region growing, and morphological parameter ex- traction, including the determination of filament skeletons. The coordinates of the skeleton pixels, given in a heliocentric system, are then converted to a more appropriate reference frame that follows the rotation of the Sun surface. In such a frame, a co-rotating filament is always located around the same position, and its skeletons (extracted from each image) are thus spatially close, forming a group of adjacent features. In a third step, the shape of each skeleton is compared with its neighbours using a curve-matching algorithm. This step will permit us to define the probability [P ] that two close filaments in the co-rotating frame are actually the same one observed on two different images. At the end, the pairs of features, for which the corresponding probability is greater than a threshold value, are associated using tracking identification indexes.On a representative sample of filaments, the good agreement between automated and manual tracking confirms the reliability of the technique to be applied on large data sets. Especially, this code is already used in the framework of the Heliophysics Integrated Observatory (HELIO) to populate a catalogue dedicated to solar and heliospheric features (HFC). An extension of this method to others filament observations, and possibly the sunspots, faculae, and coronal holes tracking can be also envisaged.
Although many methods perform well in single camera tracking, multi-camera tracking remains a challenging problem with less attention. DukeMTMC is a large-scale, well-annotated multi-camera tracking benchmark which makes great progress in this field. This report is dedicated to briefly introduce our method on DukeMTMC and show that simple hierarchical clustering with well-trained person re-identification features can get good results on this dataset.
The density-matrix renormalization group method has become a standard computational approach to the low-energy physics as well as dynamics of low-dimensional quantum systems. In this paper, we present a new set of applications, available as part of the ALPS package, that provide an efficient and flexible implementation of these methods based on a matrix-product state (MPS) representation. Our applications implement, within the same framework, algorithms to variationally find the ground state and low-lying excited states as well as simulate the time evolution of arbitrary one-dimensional and two-dimensional models. Implementing the conservation of quantum numbers for generic Abelian symmetries, we achieve performance competitive with the best codes in the community. Example results are provided for (i) a model of itinerant fermions in one dimension and (ii) a model of quantum magnetism.
362 -
Jonathan Schwarz
, Jelena Luketina
, Wojciech M. Czarnecki andn Agnieszka Grabska-Barwinska
2018
We introduce a conceptually simple and scalable framework for continual learning domains where tasks are learned sequentially. Our method is constant in the number of parameters and is designed to preserve performance on previously encountered tasks while accelerating learning progress on subsequent problems. This is achieved by training a network with two components: A knowledge base, capable of solving previously encountered problems, which is connected to an active column that is employed to efficiently learn the current task. After learning a new task, the active column is distilled into the knowledge base, taking care to protect any previously acquired skills. This cycle of active learning (progression) followed by consolidation (compression) requires no architecture growth, no access to or storing of previous data or tasks, and no task-specific parameters. We demonstrate the progress & compress approach on sequential classification of handwritten alphabets as well as two reinforcement learning domains: Atari games and 3D maze navigation.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
