Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userDeep Neural Network Representation of Density Functional Theory Hamiltonian
122
0
0.0
(
0
)
Ask ChatGPT about the research
No Arabic abstract
The marriage of density functional theory (DFT) and deep learning methods has the potential to revolutionize modern research of material science. Here we study the crucial problem of representing DFT Hamiltonian for crystalline materials of arbitrary configurations via deep neural network. A general framework is proposed to deal with the infinite dimensionality and covariance transformation of DFT Hamiltonian matrix in virtue of locality and use message passing neural network together with graph representation for deep learning. Our example study on graphene-based systems demonstrates that high accuracy ($sim$meV) and good transferability can be obtained for DFT Hamiltonian, ensuring accurate predictions of materials properties without DFT. The Deep Hamiltonian method provides a solution to the accuracy-efficiency dilemma of DFT and opens new opportunities to explore large-scale materials and physics.
rate research
Read More
Early processing of visual information takes place in the human retina. Mimicking neurobiological structures and functionalities of the retina provide a promising pathway to achieving vision sensor with highly efficient image processing. Here, we demonstrate a prototype vision sensor that operates via the gate-tunable positive and negative photoresponses of the van der Waals (vdW) vertical heterostructures. The sensor emulates not only the neurobiological functionalities of bipolar cells and photoreceptors but also the unique synaptic connectivity between bipolar cells and photoreceptors. By tuning gate voltage for each pixel, we achieve reconfigurable vision sensor for simultaneously image sensing and processing. Furthermore, our prototype vision sensor itself can be trained to classify the input images, via updating the gate voltages applied individually to each pixel in the sensor. Our work indicates that vdW vertical heterostructures offer a promising platform for the development of neural network vision sensor.
Despite their rich information content, electronic structure data amassed at high volumes in ab initio molecular dynamics simulations are generally under-utilized. We introduce a transferable high-fidelity neural network representation of such data in the form of tight-binding Hamiltonians for crystalline materials. This predictive representation of ab initio electronic structure, combined with machine-learning boosted molecular dynamics, enables efficient and accurate electronic evolution and sampling. When applied to a one-dimension charge-density wave material, carbyne, we are able to compute the spectral function and optical conductivity in the canonical ensemble. The spectral functions evaluated during soliton-antisoliton pair annihilation process reveal significant renormalization of low-energy edge modes due to retarded electron-lattice coupling beyond the Born-Oppenheimer limit. The availability of an efficient and reusable surrogate model for the electronic structure dynamical system will enable calculating many interesting physical properties, paving way to previously inaccessible or challenging avenues in materials modeling.
The technological performances of metallic compounds are largely influenced by atomic ordering. Although there is a general consensus that successful theories of metallic systems should account for the quantum nature of the electronic glue, existing non-perturbative high-temperature treatments are based on effective classical atomic Hamiltonians. We propose a solution for the above paradox and offer a fully quantum mechanical, though approximate, theory that on equal footing deals with both electrons and ions. By taking advantage of a coarse grained formulation of the density functional theory [Bruno et al., Phys. Rev. B 77, 155108 (2008)] we develop a MonteCarlo technique, based on an ab initio Hamiltonian, that allows for the efficient evaluation of finite temperature statistical averages. Calculations of the relevant thermodynamic quantities and of the electronic structures for CuZn and Ni$_3$V support that our theory provides an appropriate description of order-disorder phase transitions.
The class of the Generalized Coherent Potential Approximations (GCPA) to the Density Functional Theory (DFT) is introduced within the Multiple Scattering Theory formalism for dealing with, ordered or disordered, metallic alloys. All GCPA theories are based on a common ansatz for the kinetic part of the Hohenberg-Kohn functional and each theory of the class is specified by an external model concerning the potential reconstruction. The GCPA density functional consists of marginally coupled local contributions, does not depend on the details of the charge density and can be exactly rewritten as a function of the appropriate charge multipole moments associated with each lattice site. A general procedure based on the integration of the qV laws is described that allows for the explicit construction the same function. The coarse grained nature of the GCPA density functional implies great computational advantages and is connected with the O(N) scalability of GCPA algorithms. Moreover, it is shown that a convenient truncated series expansion of the GCPA functional leads to the Charge Excess Functional (CEF) theory [E. Bruno, L. Zingales and Y. Wang, Phys. Rev. Lett. {bf 91}, 166401 (2003)] which here is offered in a generalized version that includes multipolar interactions. CEF and the GCPA numerical results are compared with status of art LAPW full-potential density functional calculations for 62, bcc- and fcc-based, ordered CuZn alloys, in all the range of concentrations. These extensive tests show that the discrepancies between GCPA and CEF are always within the numerical accuracy of the calculations, both for the site charges and the total energies. Furthermore, GCPA and CEF very carefully reproduce the LAPW site charges and the total energy trends.
We present a numerical modeling workflow based on machine learning (ML) which reproduces the the total energies produced by Kohn-Sham density functional theory (DFT) at finite electronic temperature to within chemical accuracy at negligible computational cost. Based on deep neural networks, our workflow yields the local density of states (LDOS) for a given atomic configuration. From the LDOS, spatially-resolved, energy-resolved, and integrated quantities can be calculated, including the DFT total free energy, which serves as the Born-Oppenheimer potential energy surface for the atoms. We demonstrate the efficacy of this approach for both solid and liquid metals and compare results between independent and unified machine-learning models for solid and liquid aluminum. Our machine-learning density functional theory framework opens up the path towards multiscale materials modeling for matter under ambient and extreme conditions at a computational scale and cost that is unattainable with current algorithms.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
