BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:From Spectra to Usable Biochemical Information: Data Fusion and Da
 ta Augmentation Strategies for More Explainable Machine Learning
DTSTART:20260522T090000Z
DTEND:20260522T100000Z
DTSTAMP:20260611T204600Z
UID:indico-event-1605@indico.ifj.edu.pl
DESCRIPTION:Speakers: David Pérez-Guaita (University of Valencia)\n\nVibr
 ational spectroscopy has become a powerful tool for studying complex chemi
 cal systems\, particularly biological ones. However\, it generates large e
 mpirical datasets composed of highly overlapped spectral bands\, which req
 uire multivariate analysis through an ever-expanding range of machine lear
 ning approaches. The increasing sophistication of these models often makes
  it difficult to directly relate spectral features to meaningful biochemic
 al information. As a result\, many approaches remain “black box” model
 s\, limiting the physical interpretability of spectral bands and the relia
 bility of the extracted chemical insight. Explainable AI offers a pathway 
 to address this challenge by developing more transparent models that provi
 de both robust diagnostic tools and usable biochemical information to addr
 ess fundamental scientific questions.\nHere we present two complementary s
 trategies to enhance the interpretability of spectral machine learning mod
 els. First\, data fusion is used to correlate spectral information from di
 fferent techniques\, improving the robustness and interpretability of spec
 tral–biochemical relationships. Second\, data augmentation enables the g
 eneration of in silico spectra\, allowing virtual simulation and optimizat
 ion of experimental conditions while producing large and diverse datasets 
 that capture relevant variability. Together\, these approaches contribute 
 to more explainable\, reliable\, and data-efficient machine learning frame
 works for vibrational spectroscopy.\n\nhttps://indico.ifj.edu.pl/event/160
 5/
URL:https://indico.ifj.edu.pl/event/1605/
END:VEVENT
END:VCALENDAR