Physics-inspired Machine Learning for Materials Discovery

Description

This lecture explores the application of physics-inspired machine learning in materials discovery, focusing on atomic-scale materials modeling, finite-temperature thermodynamics, anharmonic free energies, and the role of symmetry in atomistic machine learning models. The instructor discusses the challenges in accurate electronic structure calculations, the use of machine learning to improve accuracy, and the importance of data efficiency and transferability. Various topics such as machine-learning for tensors, understanding the range of interactions, and the effectiveness of different model representations are covered, highlighting the potential of machine learning to enhance first-principles energetics in materials science.

Official source

https://mediaspace.epfl.ch/media/0_i9aylmoi

About this result

This page is automatically generated and may contain information that is not correct, complete, up-to-date, or relevant to your search query. The same applies to every other page on this website. Please make sure to verify the information with EPFL's official sources.

Related lectures (33)

Related concepts (38)

Third law of thermodynamics

The third law of thermodynamics states that the entropy of a closed system at thermodynamic equilibrium approaches a constant value when its temperature approaches absolute zero. This constant value cannot depend on any other parameters characterizing the system, such as pressure or applied magnetic field. At absolute zero (zero kelvins) the system must be in a state with the minimum possible energy. Entropy is related to the number of accessible microstates, and there is typically one unique state (called the ground state) with minimum energy.

Second law of thermodynamics

The second law of thermodynamics is a physical law based on universal experience concerning heat and energy interconversions. One simple statement of the law is that heat always moves from hotter objects to colder objects (or "downhill"), unless energy in some form is supplied to reverse the direction of heat flow. Another definition is: "Not all heat energy can be converted into work in a cyclic process." The second law of thermodynamics in other versions establishes the concept of entropy as a physical property of a thermodynamic system.

Tensor

In mathematics, a tensor is an algebraic object that describes a multilinear relationship between sets of algebraic objects related to a vector space. Tensors may map between different objects such as vectors, scalars, and even other tensors. There are many types of tensors, including scalars and vectors (which are the simplest tensors), dual vectors, multilinear maps between vector spaces, and even some operations such as the dot product.

Chemical thermodynamics

Chemical thermodynamics is the study of the interrelation of heat and work with chemical reactions or with physical changes of state within the confines of the laws of thermodynamics. Chemical thermodynamics involves not only laboratory measurements of various thermodynamic properties, but also the application of mathematical methods to the study of chemical questions and the spontaneity of processes. The structure of chemical thermodynamics is based on the first two laws of thermodynamics.

First law of thermodynamics

The first law of thermodynamics is a formulation of the law of conservation of energy, adapted for thermodynamic processes. A simple formulation is: "The total energy in a system remains constant, although it may be converted from one form to another." Another common phrasing is that "energy can neither be created nor destroyed". While there are many subtleties and implications that may be more precisely captured in more complex formulations, this is the essential principle of the First Law.