Institute of Solid State Physics EXAFS Spectroscopy Laboratory
Institute of Solid State Physics, University of Latvia
LU CFI Webmail

ENG   LV
QR code

Google Search:










Artificial Neural Network


Machine Learning is a method of data analysis that automates analytical model building using different approaches.

Artificial Neural Network (ANN) which is a type of machine learning algorithm or an information processing model consisting of elements, called artificial neurons, which are interconnected into several layers (one input layer, one or more hidden layers, and one output layer). The complex relationships between inputs and outputs are modeled during the training process using a large set of reference (experimental or theoretical) data. When trained properly, the ANN can fast generate the result based on the input data.


In our laboratory, we develop the ANNs methodology to perform the analysis of X-ray absorption data.




Example: "Neural network approach for characterizing structural transformations by x-ray absorption fine structure spectroscopy"
    (from J. Timoshenko, A. Anspoks, A. Cintins, A. Kuzmin, J. Purans, A.I. Frenkel, Phys. Rev. Lett. 120 (2018) 225502:1-6.)

Scheme of the EXAFS spectrum analysis using the artificial neural network


USEFUL LINKS

  • Wikipedia: Machine learning.
  • Wikipedia: Artificial Neural Network (ANN).

  • Neural Networks and Deep Learning - a free online book

  • TensorFlow - an open source machine learning library for research and production.
  • PyTorch - an optimized tensor library for deep learning using GPUs and CPUs.

  • CHGNet - a pretrained universal neural network potential for charge-informed atomistic modeling.
  • MLatom - a package for atomistic simulations with machine learning.
  • Quantum-Machine.org.


  • Google AI Studio.
  • Google NotebookLM.
  • OpenAI ChatGPT.
  • Microsoft Copilot.