A machine learning potential with low error in the potential energies does not guarantee good performance for the simulations. One of the reasons is that it is hard to train machine learning potentials with balanced descriptions of different PES regions, …
I have presented on March 20, 2024, the ongoing journey towards making excited-state simulations more accessible with the help of AI/ML. The video recordings and abstract of my talk at VISTA are now available online. About VISTA The bi-weekly seminar …
VISTA: Towards more accessible excited-state simulations with AI Read more »
MLatom@XACS makes AI-enhanced computational chemistry more accessible and supports both ground- and excited-state simulations with quantum mechanical methods, machine learning, and their combinations. We are happy to announce that we will release the new upgraded version of MLatom 3.3.0 that …
Surface hopping dynamics with MLatom is coming: Join online broadcast! Read more »
My review ‘AI in computational chemistry through the lens of a decade-long journey’ was published open access as an invited Feature Article in Chemical Communication. It gives a perspective on the progress of AI tools in computational chemistry through the …
AI-accelerated nonadiabatic dynamics reduces the cost of the ab initio simulations of nonlinear time-resolved spectra. We have developed a robust protocol and demonstrated its feasibility for calculating stimulated emission contributions in transient absorption pump–probe and 2D electronic spectra of pyrazine. …
The capabilities of MLatom 3 are described in the paper published in the J. Chem. Theory Comput. which was converted into an interactive online manual. See also the video overview.
Our editorial on the Special Topic “Modern semiempirical methods” was published in JCP. We overview contributions and overview the trends in the development and applications of semiempirical quantum mechanical methods. The contributions include improvements in the formalism of the method …
Editorial on the Special Topic “Modern semiempirical methods” published in JCP Read more »
Molecular dynamics simulations are widely used to study molecules and materials and lots of effort is put into making these simulations obey the physical laws. Energy conservation law is obviously one of the most important laws that MD should respect. …
Energy-conserving molecular dynamics is not energy conserving! Read more »
We have introduced a concept of 4D-spacetime atomistic AI models that learn how the molecule changes in time. We demonstrate that this concept is feasible by developing the 4D-spacetime GICnet models that directly predict the atomic coordinates of a molecule …
Surging efforts and fast progress in AI methods for photochemistry and photophysics make it difficult to track the current state of the art. We cover the recent developments in this field in the chapter on Machine learning methods in photochemistry …
Chapter “Machine Learning Methods in Photochemistry and Photophysics” Read more »