Delta-learning is slower than pure ML models but the benefit is typically an increased robustness and accuracy.

MLatom provides you not just with many lego-bricks (both ML models and quantum chemical methods) but also with the powerful and intuitive tools to glue them together in arbitrarily complex workflows!

We have developed an interpretable machine learning approach based on experimental data which predicts two-photon absorption strength instantaneously with accuracy comparable to DFT.

This work showcases the application of the block localized wave function (BLW) method to explain the impact of solvents on the strength of coordinate covalent and ionic bonds.

Our active learning protocol for accelerating surface hopping dynamics with machine learning is now available in MLatom 3.10!