A first-principle mechanism for particulate aggregation and self-assembly in stratified fluids

The ability to predict material properties without the need of resource consuming experimental efforts can immensely accelerate material and drug discovery. Although ab initio methods can be reliable and accurate in making such predictions, they are computationally too expensive at a large scale. The recent advancements in artificial intelligence and machine learning as well as availability of large quantum mechanics derived datasets enable us to train models on these datasets as benchmark and to make fast predictions on much larger datasets. The success of these machine learning models highly depends on the machine-readable fingerprints of the molecules that capture their chemical properties as well as topological information. In this work we propose a common deep learning based framework to combine different types of molecular fingerprints to enhance prediction accuracy. Graph Neural Network (GNN), Many Body Tensor Representation (MBTR) and a set of simple Molecular Descriptors (MD) were used to predict the total energies, Highest Occupied Molecular Orbital (HOMO) energies and Lowest Unoccupied Molecular Orbital (LUMO) energies of a dataset containing ~62k large organic molecules with complex aromatic rings and remarkably diverse functional groups. The results demonstrate that a combination of best performing molecular fingerprints can produce better results than the individual ones. The simple and flexible deep learning framework developed in this work can be easily adapted to incorporate other types of molecular fingerprints.

Download Full-text

Deep Learning Total Energies and Orbital Energies of Large Organic Molecules Using Hybridization of Molecular Fingerprints

10.26434/chemrxiv.12581381.v1 ◽

2020 ◽

Author(s):

Obaidur Rahaman ◽

Alessio Gagliardi

Keyword(s):

Machine Learning ◽

Deep Learning ◽

Molecular Orbital ◽

Large Scale ◽

Organic Molecules ◽

Chemical Properties ◽

Many Body ◽

Molecular Fingerprints ◽

Total Energies ◽

The Individual

The ability to predict material properties without the need of resource consuming experimental efforts can immensely accelerate material and drug discovery. Although ab initio methods can be reliable and accurate in making such predictions, they are computationally too expensive at a large scale. The recent advancements in artificial intelligence and machine learning as well as availability of large quantum mechanics derived datasets enable us to train models on these datasets as benchmark and to make fast predictions on much larger datasets. The success of these machine learning models highly depends on the machine-readable fingerprints of the molecules that capture their chemical properties as well as topological information. In this work we propose a common deep learning based framework to combine different types of molecular fingerprints to enhance prediction accuracy. Graph Neural Network (GNN), Many Body Tensor Representation (MBTR) and a set of simple Molecular Descriptors (MD) were used to predict the total energies, Highest Occupied Molecular Orbital (HOMO) energies and Lowest Unoccupied Molecular Orbital (LUMO) energies of a dataset containing ~62k large organic molecules with complex aromatic rings and remarkably diverse functional groups. The results demonstrate that a combination of best performing molecular fingerprints can produce better results than the individual ones. The simple and flexible deep learning framework developed in this work can be easily adapted to incorporate other types of molecular fingerprints.

Download Full-text

THE TRANSPORT COEFFICIENTS OF LARGE SCALE NUCLEAR COLLECTIVE MOTION

International Journal of Modern Physics E ◽

10.1142/s0218301308009550 ◽

2008 ◽

Vol 17 (01) ◽

pp. 60-71

Author(s):

FEDIR IVANYUK

Keyword(s):

Large Scale ◽

Collective Motion ◽

Transport Coefficients ◽

Harmonic Approximation ◽

Linear Response Theory ◽

Many Body ◽

Collective Variables ◽

Angular Rotation ◽

Shell Effects ◽

Deformation Parameters

An approach to the description of slow collective motion at finite excitations based on the linear response theory and locally harmonic approximation is discussed. The method relies on the assumption that the nuclear many-body Hamiltonian can be approximated by a one-body Hamiltonian which in parametric way depends on the collective variables (deformation parameters) and allows for the account of the shell effects, pairing correlation, angular rotation. Special attention is paid to the removal of spurious contributions caused by the violation of the conservation laws. As an application of the theory, the microscopic transport coefficient are used for the description of reaction 18O + 208 Pb within the Langevin approach.

Download Full-text

Intelligent Control of Swarm Robotics Employing Biomimetic Deep Learning

Machines ◽

10.3390/machines9100236 ◽

2021 ◽

Vol 9 (10) ◽

pp. 236

Author(s):

Haoxiang Zhang ◽

Lei Liu

Keyword(s):

Deep Learning ◽

Large Scale ◽

Swarm Robotics ◽

Collective Motion ◽

Control Method ◽

Interaction Model ◽

Pairwise Interaction ◽

Selection Strategy ◽

Neighbor Selection ◽

The Individual

The collective motion of biological species has robust and flexible characteristics. Since the individual of the biological group interacts with other neighbors asymmetrically, which means the pairwise interaction presents asymmetrical characteristics during the collective motion, building the model of the pairwise interaction of the individual is still full of challenges. Based on deep learning (DL) technology, experimental data of the collective motion on Hemigrammus rhodostomus fish are analyzed to build an individual interaction model with multi-parameter input. First, a Deep Neural Network (DNN) structure for pairwise interaction is designed. Then, the interaction model is obtained by means of DNN proper training. We propose a novel key neighbor selection strategy, which is called the Largest Visual Pressure Selection (LVPS) method, to deal with multi-neighbor interaction. Based on the information of the key neighbor identified by LVPS, the individual uses the properly trained DNN model for the pairwise interaction. Compared with other key neighbor selection strategies, the statistical properties of the collective motion simulated by our proposed DNN model are more consistent with those of fish experiments. The simulation shows that our proposed method can extend to large-scale group collective motion for aggregation control. Thereby, the individual can take advantage of quite limited local information to collaboratively achieve large-scale collective motion. Finally, we demonstrate swarm robotics collective motion in an experimental platform. The proposed control method is simple to use, applicable for different scales, and fast for calculation. Thus, it has broad application prospects in the fields of multi-robotics control, intelligent transportation systems, saturated cluster attacks, and multi-agent logistics, among other fields.

Download Full-text

Progress in defining the biological causes of schizophrenia

Expert Reviews in Molecular Medicine ◽

10.1017/s1462399411001955 ◽

2011 ◽

Vol 13 ◽

Cited By ~ 18

Author(s):

Benjamin Pickard

Keyword(s):

Risk Factors ◽

Large Scale ◽

Biological Process ◽

Biological Information ◽

Data Sets ◽

Complex Interplay ◽

Disease Biomarkers ◽

Rational Development ◽

Nervous System Function ◽

The Individual

Schizophrenia is a common mental illness resulting from a complex interplay of genetic and environmental risk factors. Establishing its primary molecular and cellular aetiopathologies has proved difficult. However, this is a vital step towards the rational development of useful disease biomarkers and new therapeutic strategies. The advent and large-scale application of genomic, transcriptomic, proteomic and metabolomic technologies are generating data sets required to achieve this goal. This discovery phase, typified by its objective and hypothesis-free approach, is described in the first part of the review. The accumulating biological information, when viewed as a whole, reveals a number of biological process and subcellular locations that contribute to schizophrenia causation. The data also show that each technique targets different aspects of central nervous system function in the disease state. In the second part of the review, key schizophrenia candidate genes are discussed more fully. Two higher-order processes – adult neurogenesis and inflammation – that appear to have pathological relevance are also described in detail. Finally, three areas where progress would have a large impact on schizophrenia biology are discussed: deducing the causes of schizophrenia in the individual, explaining the phenomenon of cross-disorder risk factors, and distinguishing causative disease factors from those that are reactive or compensatory.

Download Full-text

Models, Practices and Methods of Reading: Evolution in Time and Space

Bibliotekovedenie [Library and Information Science (Russia)] ◽

10.25281/0869-608x-2009-0-1-59-64 ◽

2009 ◽

pp. 59-64

Author(s):

Yulia P. Melentyeva

Keyword(s):

Large Scale ◽

Social Groups ◽

National Program ◽

Time And Space ◽

Lack Of Information ◽

The Individual

In recent years as public in general and specialist have been showing big interest to the matters of reading. According to discussion and launch of the “Support and Development of Reading National Program”, many Russian libraries are organizing the large-scale events like marathons, lecture cycles, bibliographic trainings etc. which should draw attention of different social groups to reading. The individual forms of attraction to reading are used much rare. To author’s mind the main reason of such an issue has to be the lack of information about forms and methods of attraction to reading.

Download Full-text

Double Precision Is Not Needed for Many-Body Calculations: New Conventional Wisdom

10.26434/chemrxiv.6104804.v1 ◽

2018 ◽

Author(s):

Pavel Pokhilko ◽

Evgeny Epifanovsky ◽

Anna I. Krylov

Keyword(s):

Large Scale ◽

Computation Time ◽

Coupled Cluster ◽

Double Precision ◽

Many Body ◽

Single Precision ◽

Parallel Performance ◽

Point Representation ◽

Electron Repulsion Integrals ◽

Cluster Methods

Using single precision floating point representation reduces the size of data and computation time by a factor of two relative to double precision conventionally used in electronic structure programs. For large-scale calculations, such as those encountered in many-body theories, reduced memory footprint alleviates memory and input/output bottlenecks. Reduced size of data can lead to additional gains due to improved parallel performance on CPUs and various accelerators. However, using single precision can potentially reduce the accuracy of computed observables. Here we report an implementation of coupled-cluster and equation-of-motion coupled-cluster methods with single and double excitations in single precision. We consider both standard implementation and one using Cholesky decomposition or resolution-of-the-identity of electron-repulsion integrals. Numerical tests illustrate that when single precision is used in correlated calculations, the loss of accuracy is insignificant and pure single-precision implementation can be used for computing energies, analytic gradients, excited states, and molecular properties. In addition to pure single-precision calculations, our implementation allows one to follow a single-precision calculation by clean-up iterations, fully recovering double-precision results while retaining significant savings.

Download Full-text

REGULAR STRUCTURE OF LOW-LYING STATES FOR ATOMIC NUCLEI UNDER RANDOM INTERACTIONS

International Journal of Modern Physics E ◽

10.1142/s021830130801194x ◽

2008 ◽

Vol 17 (supp01) ◽

pp. 304-317

Author(s):

Y. M. ZHAO

Keyword(s):

Ground State ◽

Collective Motion ◽

Regular Structure ◽

Positive Parity ◽

Many Body ◽

Random Interactions ◽

Atomic Nuclei ◽

Many Body Systems

In this paper we review regularities of low-lying states for many-body systems, in particular, atomic nuclei, under random interactions. We shall discuss the famous problem of spin zero ground state dominance, positive parity dominance, collective motion, odd-even staggering, average energies, etc., in the presence of random interactions.

Download Full-text

Entangling Lattice-Trapped Bosons with a Free Impurity: Impact on Stationary and Dynamical Properties

Entropy ◽

10.3390/e23030290 ◽

2021 ◽

Vol 23 (3) ◽

pp. 290

Author(s):

Maxim Pyzh ◽

Kevin Keiler ◽

Simeon I. Mistakidis ◽

Peter Schmelcher

Keyword(s):

Structural Changes ◽

Many Body ◽

Wide Range ◽

Entropy Measures ◽

Particle Level ◽

The Many ◽

The One ◽

Tunneling Dynamics ◽

The Individual ◽

Trapped Bosons

We address the interplay of few lattice trapped bosons interacting with an impurity atom in a box potential. For the ground state, a classification is performed based on the fidelity allowing to quantify the susceptibility of the composite system to structural changes due to the intercomponent coupling. We analyze the overall response at the many-body level and contrast it to the single-particle level. By inspecting different entropy measures we capture the degree of entanglement and intraspecies correlations for a wide range of intra- and intercomponent interactions and lattice depths. We also spatially resolve the imprint of the entanglement on the one- and two-body density distributions showcasing that it accelerates the phase separation process or acts against spatial localization for repulsive and attractive intercomponent interactions, respectively. The many-body effects on the tunneling dynamics of the individual components, resulting from their counterflow, are also discussed. The tunneling period of the impurity is very sensitive to the value of the impurity-medium coupling due to its effective dressing by the few-body medium. Our work provides implications for engineering localized structures in correlated impurity settings using species selective optical potentials.

Download Full-text