Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Generative AI in Spectroscopy

Nath, Debarati ; Ghosh, Sukanya ; Bhattacharya, Sayantani ; Hiremath, Praveenkumar LU and Deb, Debajit (2025) In Progress in Optical Science and Photonics 36. p.165-186
Abstract

Generative AI with spectroscopy is an innovative approach that combines artificial intelligence techniques with spectroscopic analysis, enabling researchers to extract valuable insights from complex spectral data more efficiently and accurately. AI can create high-precision synthetic spectra, enhance signal-to-noise ratios, and support spectral reconstruction by employing deep learning techniques like Generative Adversarial Networks (GANs), Graph Neural Networks (GNN), and Variational Autoencoders (VAEs). GenAI models facilitate the simulation of spectra from molecular structures, solving the forward design, whereas reasoning-driven models address the inverse design by predicting molecular structures with greater accuracy. GenAI can be... (More)

Generative AI with spectroscopy is an innovative approach that combines artificial intelligence techniques with spectroscopic analysis, enabling researchers to extract valuable insights from complex spectral data more efficiently and accurately. AI can create high-precision synthetic spectra, enhance signal-to-noise ratios, and support spectral reconstruction by employing deep learning techniques like Generative Adversarial Networks (GANs), Graph Neural Networks (GNN), and Variational Autoencoders (VAEs). GenAI models facilitate the simulation of spectra from molecular structures, solving the forward design, whereas reasoning-driven models address the inverse design by predicting molecular structures with greater accuracy. GenAI can be integrated with spectroscopy to achieve various tasks such as improvement of spectral data analysis, rapid advancement in material discovery, automation, closed-loop optimization, etc. These innovations allow faster and more precise material identification, and real-time spectral prediction across diverse fields such as pharmaceuticals, environmental monitoring, and materials science. Despite challenges related to data quality, model interpretability, and computational demands, integrating Generative AI with spectroscopy offers significant potential for driving advancements in scientific research and industrial applications.

(Less)
Please use this url to cite or link to this publication:
author
; ; ; and
organization
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
keywords
Forward design, Generative adverrial neworks, Inverse design, Molecular structure, Spectroscopy
host publication
Progress in Optical Science and Photonics
series title
Progress in Optical Science and Photonics
volume
36
pages
22 pages
publisher
Springer
external identifiers
  • scopus:105020958704
ISSN
2363-510X
2363-5096
DOI
10.1007/978-981-95-1561-5_6
language
English
LU publication?
yes
additional info
Publisher Copyright: © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.
id
c3a6b389-ba30-493f-9465-ec87686880df
date added to LUP
2026-01-13 13:43:19
date last changed
2026-01-13 13:43:45
@inbook{c3a6b389-ba30-493f-9465-ec87686880df,
  abstract     = {{<p>Generative AI with spectroscopy is an innovative approach that combines artificial intelligence techniques with spectroscopic analysis, enabling researchers to extract valuable insights from complex spectral data more efficiently and accurately. AI can create high-precision synthetic spectra, enhance signal-to-noise ratios, and support spectral reconstruction by employing deep learning techniques like Generative Adversarial Networks (GANs), Graph Neural Networks (GNN), and Variational Autoencoders (VAEs). GenAI models facilitate the simulation of spectra from molecular structures, solving the forward design, whereas reasoning-driven models address the inverse design by predicting molecular structures with greater accuracy. GenAI can be integrated with spectroscopy to achieve various tasks such as improvement of spectral data analysis, rapid advancement in material discovery, automation, closed-loop optimization, etc. These innovations allow faster and more precise material identification, and real-time spectral prediction across diverse fields such as pharmaceuticals, environmental monitoring, and materials science. Despite challenges related to data quality, model interpretability, and computational demands, integrating Generative AI with spectroscopy offers significant potential for driving advancements in scientific research and industrial applications.</p>}},
  author       = {{Nath, Debarati and Ghosh, Sukanya and Bhattacharya, Sayantani and Hiremath, Praveenkumar and Deb, Debajit}},
  booktitle    = {{Progress in Optical Science and Photonics}},
  issn         = {{2363-510X}},
  keywords     = {{Forward design; Generative adverrial neworks; Inverse design; Molecular structure; Spectroscopy}},
  language     = {{eng}},
  pages        = {{165--186}},
  publisher    = {{Springer}},
  series       = {{Progress in Optical Science and Photonics}},
  title        = {{Generative AI in Spectroscopy}},
  url          = {{http://dx.doi.org/10.1007/978-981-95-1561-5_6}},
  doi          = {{10.1007/978-981-95-1561-5_6}},
  volume       = {{36}},
  year         = {{2025}},
}