BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Hands on Machine Learning
DTSTART:20261104T130000Z
DTEND:20261204T160000Z
DTSTAMP:20261114T052600Z
UID:indico-event-4184@indico.ijs.si
CONTACT:sbhattacharyya@ung.si\;events@slaif.si
DESCRIPTION:Speakers: Saptashwa Bhattacharyya (UNG (Center for Cosmology a
 nd Astrophysics))\n\nIzvajalec / Course provider: University of Nova Goric
 a (UNG)Predavatelji / Instructors: G. Zaharijaš and S. Bhattacharyya\nSho
 rt course description\nThis course is a hands-on microcredential focused o
 n building robust ML analysis pipelines for real-world data. Participants 
 progress from data preparation\, regression and classification through mod
 el evaluation and cross-validation to deep neural networks and modern gene
 rative models. The course combines lectures with practical sessions. The c
 ourse credits will be awarded to candidates who fulfill the attendance req
 uirement (in person/online) and submit a short report/presentation detaili
 ng their methodology and practical application.\nObjectives and competenci
 es\n\nBuild a robust data-analysis pipeline using machine-learning methods
 .\nApply the techniques to a broad range of real-world data\, including ta
 bular and unstructured data such as images.\nDevelop practical experience 
 through hands-on exercises and a final project.\n\nIntended learning outco
 mes\n\nHandle and process data before applying machine-learning methods.\n
 Apply machine-learning algorithms to practical datasets.\nSelect\, evaluat
 e and compare ML models using appropriate performance metrics and validati
 on procedures.\nApply the techniques learned in the course to a participan
 t-selected use case.\n\nCourse schedule\nTen sessions\, each 14:00–17:00
 :\n04/11/2026 Introduction to regression and classification06/11/2026 Perf
 ormance metrics and evaluation11/11/2026 Building an analysis pipeline and
  cross-validation13/11/2026 Introduction to DNNs: backpropagation and Dens
 eNet18/11/2026 Analysis with DenseNet: tabular and image data20/11/2026 In
 troduction to CNNs and different architectures25/11/2026 Image classificat
 ion and segmentation with CNNs27/11/2026 Variational Autoencoder and GAN02
 /12/2026 Probabilistic diffusion models04/12/2026 Generation and evaluatio
 n of images\n \nLearning and teaching methods\n\nLectures combined with h
 ands-on sessions.\nExercises applying different methods and evaluating the
  performance of ML models.\nFinal project: construction of a complete ML a
 nalysis pipeline for a particular task using a dataset chosen by the parti
 cipant.\n\nAssessment\n\nClass attendance: 35%\nFinal project and presenta
 tion: 65%\n\nReadings\n\nDeep Learning (Y. Bengio et al.) – available on
 line.\nStanford machine-learning course notes – available online.\nUnder
 standing Deep Learning (S. Prince) – available online.\n\n\nhttps://indi
 co.ijs.si/event/4184/
IMAGE;VALUE=URI:https://indico.ijs.si/event/4184/logo-3560331715.png
LOCATION:Amphitheater (Ajdovščina)
URL:https://indico.ijs.si/event/4184/
END:VEVENT
END:VCALENDAR