scholarly journals SpookyNet: Learning force fields with electronic degrees of freedom and nonlocal effects

2021 ◽  
Vol 12 (1) ◽  
Author(s):  
Oliver T. Unke ◽  
Stefan Chmiela ◽  
Michael Gastegger ◽  
Kristof T. Schütt ◽  
Huziel E. Sauceda ◽  
...  
Keyword(s):  
Quantum Chemistry ◽  
Network Architecture ◽  
Degrees Of Freedom ◽  
State Of The Art ◽  
Force Fields ◽  
Conformational Space ◽  
Total Charge ◽  
Data Sets ◽  
Nonlocal Effects ◽  
Current State

AbstractMachine-learned force fields combine the accuracy of ab initio methods with the efficiency of conventional force fields. However, current machine-learned force fields typically ignore electronic degrees of freedom, such as the total charge or spin state, and assume chemical locality, which is problematic when molecules have inconsistent electronic states, or when nonlocal effects play a significant role. This work introduces SpookyNet, a deep neural network for constructing machine-learned force fields with explicit treatment of electronic degrees of freedom and nonlocality, modeled via self-attention in a transformer architecture. Chemically meaningful inductive biases and analytical corrections built into the network architecture allow it to properly model physical limits. SpookyNet improves upon the current state-of-the-art (or achieves similar performance) on popular quantum chemistry data sets. Notably, it is able to generalize across chemical and conformational space and can leverage the learned chemical insights, e.g. by predicting unknown spin states, thus helping to close a further important remaining gap for today’s machine learning models in quantum chemistry.

scholarly journals Critical Aspects of Person Counting and Density Estimation

2021 ◽  
Vol 7 (2) ◽  
pp. 21
Author(s):  
Roland Perko ◽  
Manfred Klopschitz ◽  
Alexander Almer ◽  
Peter M. Roth
Keyword(s):  
Density Estimation ◽  
Network Architecture ◽  
Reference Data ◽  
State Of The Art ◽  
Limit State ◽  
Ground Truth ◽  
Data Sets ◽  
Ground Truth Generation ◽  
Baseline Approach ◽  
Critical Aspects

Many scientific studies deal with person counting and density estimation from single images. Recently, convolutional neural networks (CNNs) have been applied for these tasks. Even though often better results are reported, it is often not clear where the improvements are resulting from, and if the proposed approaches would generalize. Thus, the main goal of this paper was to identify the critical aspects of these tasks and to show how these limit state-of-the-art approaches. Based on these findings, we show how to mitigate these limitations. To this end, we implemented a CNN-based baseline approach, which we extended to deal with identified problems. These include the discovery of bias in the reference data sets, ambiguity in ground truth generation, and mismatching of evaluation metrics w.r.t. the training loss function. The experimental results show that our modifications allow for significantly outperforming the baseline in terms of the accuracy of person counts and density estimation. In this way, we get a deeper understanding of CNN-based person density estimation beyond the network architecture. Furthermore, our insights would allow to advance the field of person density estimation in general by highlighting current limitations in the evaluation protocols.

scholarly journals A Systematic Literature Review on Military Software Defined Networks

2018 ◽  
Vol 10 (9) ◽  
pp. 88 ◽  
Author(s):  
Vasileios Gkioulos ◽  
Håkon Gunleifsen ◽  
Goitom Weldehawaryat
Keyword(s):  
Literature Review ◽  
Systematic Literature Review ◽  
Network Architecture ◽  
State Of The Art ◽  
Future Research ◽  
Software Defined Networks ◽  
Application Domain ◽  
Research Areas ◽  
Current State ◽  
Active Research

Software Defined Networking (SDN) is an evolving network architecture paradigm that focuses on the separation of control and data planes. SDN receives increasing attention both from academia and industry, across a multitude of application domains. In this article, we examine the current state of obtained knowledge on military SDN by conducting a systematic literature review (SLR). Through this work, we seek to evaluate the current state of the art in terms of research tracks, publications, methods, trends, and most active research areas. Accordingly, we utilize these findings for consolidating the areas of past and current research on the examined application domain, and propose directions for future research.

scholarly journals Native Language Identification With Classifier Stacking and Ensembles

2018 ◽  
Vol 44 (3) ◽  
pp. 403-446 ◽  
Author(s):  
Shervin Malmasi ◽  
Mark Dras
Keyword(s):  
State Of The Art ◽  
Native Language ◽  
Ensemble Methods ◽  
Large Data ◽  
Language Identification ◽  
Large Data Sets ◽  
Data Sets ◽  
Classification Models ◽  
Multiple Classifiers ◽  
Current State

Ensemble methods using multiple classifiers have proven to be among the most successful approaches for the task of Native Language Identification (NLI), achieving the current state of the art. However, a systematic examination of ensemble methods for NLI has yet to be conducted. Additionally, deeper ensemble architectures such as classifier stacking have not been closely evaluated. We present a set of experiments using three ensemble-based models, testing each with multiple configurations and algorithms. This includes a rigorous application of meta-classification models for NLI, achieving state-of-the-art results on several large data sets, evaluated in both intra-corpus and cross-corpus modes.

Multivariate phase combination improves automated crystallographic model building

2010 ◽  
Vol 66 (7) ◽  
pp. 783-788 ◽  
Author(s):  
Pavol Skubák ◽  
Willem-Jan Waterreus ◽  
Navraj S. Pannu
Keyword(s):  
Model Building ◽  
State Of The Art ◽  
Standard Technique ◽  
Real Data ◽  
Data Sets ◽  
Macromolecular Crystallography ◽  
Current State ◽  
Phase Combination ◽  
Density Map ◽  
Electron Density Map

Density modification is a standard technique in macromolecular crystallography that can significantly improve an initial electron-density map. To obtain optimal results, the initial and density-modified map are combined. Current methods assume that these two maps are independent and propagate the initial map information and its accuracy indirectly through previously determined coefficients. A multivariate equation has been derived that no longer assumes independence between the initial and density-modified map, considers the observed diffraction data directly and refines the errors that can occur in a single-wavelength anomalous diffraction experiment. The equation has been implemented and tested on over 100 real data sets. The results are dramatic: the method provides significantly improved maps over the current state of the art and leads to many more structures being built automatically.

scholarly journals Euler diagrams drawn with ellipses area-proportionally (Edeap)

2021 ◽  
Vol 22 (1) ◽  
Author(s):  
Michael Wybrow ◽  
Peter Rodgers ◽  
Fadi K. Dib
Keyword(s):  
State Of The Art ◽  
Data Sets ◽  
Microarray Experiments ◽  
Current State ◽  
Gene Ontology Database ◽  
Euler Diagrams ◽  
Twitter Data ◽  
Extensive Evaluation ◽  
Using Data ◽  
The Web

AbstractBackgroundArea-proportional Euler diagrams are frequently used to visualize data from Microarray experiments, but are also applied to a wide variety of other data from biosciences, social networks and other domains.ResultsThis paper details Edeap, a new simple, scalable method for drawing area-proportional Euler diagrams with ellipses. We use a search-based technique optimizing a multi-criteria objective function that includes measures for both area accuracy and usability, and which can be extended to further user-defined criteria. The Edeap software is available for use on the web, and the code is open source. In addition to describing our system, we present the first extensive evaluation of software for producing area-proportional Euler diagrams, comparing Edeap to the current state-of-the-art; circle-based method, venneuler, and an alternative ellipse-based method, eulerr.ConclusionsOur evaluation—using data from the Gene Ontology database via GoMiner, Twitter data from the SNAP database, and randomly generated data sets—shows an ordering for accuracy (from best to worst) of eulerr, followed by Edeap and then venneuler. In terms of runtime, the results are reversed with venneuler being the fastest, followed by Edeap and finally eulerr. Regarding scalability, eulerr cannot draw non-trivial diagrams beyond 11 sets, whereas no such limitation is present in Edeap or venneuler, both of which draw diagrams up to the tested limit of 20 sets.

scholarly journals Supervised and Unsupervised Neural Approaches to Text Readability

2021 ◽  
Vol 47 (1) ◽  
pp. 141-179
Author(s):  
Matej Martinc ◽  
Senja Pollak ◽  
Marko Robnik-Šikonja
Keyword(s):  
State Of The Art ◽  
Comprehensive Analysis ◽  
Language Models ◽  
Feature Engineering ◽  
Data Sets ◽  
Data Set ◽  
Systematic Comparison ◽  
Current State ◽  
Text Readability ◽  
Unsupervised Approach

Abstract We present a set of novel neural supervised and unsupervised approaches for determining the readability of documents. In the unsupervised setting, we leverage neural language models, whereas in the supervised setting, three different neural classification architectures are tested. We show that the proposed neural unsupervised approach is robust, transferable across languages, and allows adaptation to a specific readability task and data set. By systematic comparison of several neural architectures on a number of benchmark and new labeled readability data sets in two languages, this study also offers a comprehensive analysis of different neural approaches to readability classification. We expose their strengths and weaknesses, compare their performance to current state-of-the-art classification approaches to readability, which in most cases still rely on extensive feature engineering, and propose possibilities for improvements.

GA2RM: A GA-Based Action Rule Mining Method

2021 ◽  
pp. 2150012
Author(s):  
Shervin Hashemi ◽  
Pirooz Shamsinejad
Keyword(s):  
State Of The Art ◽  
Synthetic Data ◽  
Search Space ◽  
Data Sets ◽  
Mining Method ◽  
Classification Rules ◽  
Classification Methods ◽  
Rule Mining ◽  
Current State ◽  
Traditional Classification

Action Mining is a subfield of Data Mining that tries to extract actions from traditional data sets. Action Rule is a type of rule that suggests some changes in its consequent part. Extracting action rules from data has been one of the research interests in recent years. Current state-of-the-art action rule mining methods like DEAR typically take classification rules as their input; Since traditional classification methods have been designed for prediction and not for manipulation, therefore extracting action rules directly from data can result in more valuable action rules. Here, we have proposed a method to generate action rules directly from data. To tackle the problem of huge search space of action rules, a Genetic Algorithm has been devised. Different metrics have been defined for investigating the effectiveness of our proposed method and a large number of experiments have been done on real and synthetic data sets. The results show that our method can find from 20% to 10 times more interesting (in case of support and confidence) action rules in comparison with its competitors.

scholarly journals Natural product biosynthesis: What's next? An introduction to the JBC Reviews Thematic Series

2019 ◽  
Vol 295 (2) ◽  
pp. 335-336
Author(s):  
Satish K. Nair ◽  
Joseph M. Jez
Keyword(s):  
Natural Product ◽  
State Of The Art ◽  
Data Sets ◽  
Specialized Metabolites ◽  
Current State ◽  
Modified Peptides ◽  
Thematic Review ◽  
Global Threat ◽  
Derivatives Of ◽  
Academic Sector

The diversity of natural products not only fascinates us intellectually, but also provides an armamentarium against the microbes that threaten our health. The increased prevalence of pathogens that are resistant to one or more classes of available medicines continues to be a growing global threat. As drug-resistant pathogens erode the effectiveness of the current reserve of antibiotics and antifungals, methodological advances open additional avenues for discovery of new classes of drugs, as well as novel derivatives of existing (and proven) classes of compounds. In this Thematic Review Series, we aim to provide a snapshot of the current state of the art in natural product discovery. The reviews in this series encapsulate convergent approaches toward the identification of different classes of primary and specialized metabolites, including nonribosomal peptides, polyketides, and ribosomally synthesized and post-translationally modified peptides, from all kingdoms of life. Traction in unraveling new and diverse classes of molecules has come largely from the academic sector, which ensures availability of methods and data sets. Such knowledge is needed to thwart serious threats to human health and calls to mind the proverb praemonitus praemunitus (forewarned is forearmed).

scholarly journals Seed Selection Strategies for Overlap Detection

2018 ◽  
Author(s):  
Jonathan Teutenberg
Keyword(s):  
State Of The Art ◽  
Data Sets ◽  
Seed Selection ◽  
Selection Strategies ◽  
Current State ◽  
Long Read ◽  
Selection Of

AbstractThe current state-of-the-art assemblers of long, error-prone reads rely on detecting all-vs-all overlaps within the set of reads with overlaps represented by a sparse selection of short subsequences or “seeds”. Though the quality of selection of these seeds can impact both accuracy and speed of overlap detection, existing algorithms do little more than ignore over-represented seeds. Here we propose several more informed seed selection strategies to improve precision and recall of overlaps. These strategies are evaluated against real long-read data sets with a range of fixed seed sizes. We show that these strategies substantially improve the utility of individual seeds over uninformed selection.

scholarly journals Experimental and Numerical Investigations of Pipeline with Resonator

2019 ◽  
Vol 23 (1) ◽  
pp. 17-22
Author(s):  
Tomasz Pałczyński ◽  
K. Kantyka
Keyword(s):  
Degrees Of Freedom ◽  
Pressure Pulsation ◽  
State Of The Art ◽  
Dynamic Properties ◽  
Experimental Investigations ◽  
Test Rig ◽  
Additional Degree ◽  
Nonlinear Properties ◽  
System Building ◽  
Current State

Abstract This article presents the current state of the art regarding the use resonators in straight pipes. There is considerable need to control and reduce pressure pulsation in pipelines supplied with pulsating flows. The use of a Helmholz resonator introduces an additional degree of freedom to the analysed dynamic system. Building on previous experimental investigations by the authors, which identified the nonlinear properties of straight pipes supplied with pulsating flows, this study describes an experimental test rig, measurement methods and mechanical analogies for one (1DOF) and two (2DOF) degrees of freedom. The results are presented in the form of a 3D map of amplitude-frequency characteristics, as a function of the resonator volume determined by piston height. The dynamic properties of the described system are presented as amplitude-phase characteristics, based on a comparison of the numerical and experimental results.

Sign in / Sign up

Export Citation Format

Share Document