Kernel regression methods for prediction of materials properties: Recent developments

Chemical Physics Reviews, Год журнала: 2025, Номер 6(1)

Опубликована: Фев. 13, 2025

Machine learning (ML) is increasingly used in chemical physics and materials science. One major area of thrust machine properties molecules solid from descriptors composition structure. Recently, kernel regression methods various flavors—such as ridge regression, Gaussian process support vector machine—have attracted attention such applications. Kernel allow benefiting simultaneously the advantages linear regressions superior expressive power nonlinear kernels. In many applications, are high-dimensional feature spaces, where sampling with training data bound to be sparse effects specific spaces significantly affect performance method. We review recent applications kernel-based for prediction structure related purposes. discuss methodological aspects including choices kernels appropriate different dimensionality, ways balance reliability model data. also regression-based hybrid ML approaches.

Язык: Английский

---

A Geometric Berry Phase Angle Induced in Im-3m H3S at 200 GPa by Ultra-Fast Laser Pulses

Symmetry, Год журнала: 2025, Номер 17(2), С. 299 - 299

Опубликована: Фев. 16, 2025

We investigated Im-3m H3S at 200 GPa, a pressure regime where crystalline is widely considered to be superconductor. Simulated circularly polarized 10 femtosecond (fs) laser pulses were applied and we quantified the effects on electron dynamics both during application of ultra-fast pulse 5.0 fs after was switched off. In addition, carrier-envelope phase (CEP) angle ϕ, which quantifies relationship between time-varying direction electric (E)-field amplitude envelope, employed control time evolution wavefunction ψ(r). This undertaken for first Next Generation Quantum Theory Atoms in Molecules (NG-QTAIM) solid state. Ultra-fast phenomena related superconductivity are discovered form geometric Berry associated with H--H bonding addition very high values chirality–helicity function that correspond normally found chiral molecules. Future applications discussed, including spin selective high-temperature organic superconductors phonons do not play significant role.

Язык: Английский

---

The Common Concept of Anticooperativity Among Molecules Is Fundamentally Flawed, Based on Novel and Unified Molecular-Wide and Electron Density (MOWeD) Concept of Chemical Bonding

Molecules, Год журнала: 2025, Номер 30(9), С. 1944 - 1944

Опубликована: Апрель 27, 2025

A non-linear (non-additive) increase in stability of hexamers follows an the total number (i) aad (a double proton acceptor) plus add donor) waters commonly linked with anticooperativity and (ii) intermolecularly delocalized electrons (intermolNdeloc) 3D space occupied by a hexamer. Subsequently, we obtained nearly perfect linear correlation between cluster intermolNdeloc. Individual water molecules that act as either or delocalize largest throughout cluster; are involved strongest attractive, hence energy-stabilizing intermolecular interaction remaining five waters; (iii) have most significant quantum component energy (iv) relative to six non-interacting molecules, stabilize hexamer most, quantified purposely derived mol-FAMSEC term. Clearly, all-body approach used unified, molecular-wide electron density (MOWeD)-based concept chemical bonding contradicts accepted view hexamers. Consequently, propose here general definition cooperativity should be applicable any n-membered molecular cluster, namely quantifiable, classical physics- quantum-based phenomenon is synonymous delocalization electrons, leading individual on formation.

Язык: Английский

---

Weak Interactions and its Topological Characteristics in (R)-3-Quinuclidinol Crystals: a Quantum Crystallography Approach

Journal of Structural Chemistry, Год журнала: 2025, Номер 66(4), С. 669 - 677

Опубликована: Апрель 1, 2025

Язык: Английский

---