BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:PHAROS Training Series - Course 14 "Machine & Deep Learning for Ti
 me Series forecasting"
DTSTART;VALUE=DATE-TIME:20260915T080000Z
DTEND;VALUE=DATE-TIME:20260915T110000Z
DTSTAMP;VALUE=DATE-TIME:20260815T001746Z
UID:indico-event-219@events.grnet.gr
DESCRIPTION:\n\nPHAROS AI Factory announces the 14th Course of its Trainin
 g Series\, under the title "Machine & Deep Learning for Time Series foreca
 sting"\, under the topic Deep Learning\, held online via Zoom. \n\nDate
 : September 15th\, 2026\, at 11:00 EEST \n\nLocation: Online via Zoom\n\
 nPresentation Language: Greek\n\nCourse Description: This hands-on sessio
 n works through a complete forecasting pipeline in Python related to time-
 series data. Participants begin with preprocessing (handling missing value
 s\, outliers\, stationarity\, and feature engineering)\, then move through
  classical machine learning models\, gradient-boosted trees\, and deep lea
 rning architectures including LSTMs\, temporal convolutional networks\, an
 d Transformer-based models such as PatchTST. The course is built around re
 alistic use cases\, working with live coding alongside participants.\n\nAu
 dience:  ML Engineers\, Data Scientists\, Academic Researchers\, Business
  Owners \n\nPrerequisites: Python \n\nLearning Objectives: \n\n\n	Prepar
 e time series data for supervised learning (handling irregular sampling\, 
 missing values and outliers — while avoiding the data leakage patterns t
 hat invalidate most forecasting results).\n	Select appropriate baselines a
 nd evaluation metrics\, and apply walk-forward validation rather than stan
 dard cross-validation.\n	Build and compare classical\, gradient-boosting a
 nd deep learning forecasting models\, and judge when the added complexity 
 of deep learning is justified.\n	Recognise the strengths and limitations o
 f current architectures\, from LSTMs and TCNs to Transformer-based and fou
 ndation models.\n\n\nInstructor's profile:\n\n\n	Efterpi Paraskevoulakou\,
  UPRC\n\n\nEfterpi Paraskevoulakou is a researcher and practitioner worki
 ng at the intersection of applied machine learning and large-scale systems
 . She holds a PhD from the University of Piraeus (Department of Digital Sy
 stems) on AI-driven resource and service management across the cloud-edge 
 continuum\, with an M.Sc. in Big Data Analytics and a B.Sc. in Economics f
 rom the same institution. She has published in leading peer-reviewed journ
 als and international conferences\, including IEEE Transactions on Network
  and Service Management and Future Generation Computer Systems\, and recei
 ved a Best Paper Award at CCWC 2024. She has also participated in several 
 EU-funded research projects focused on AI-driven orchestration in cloud-ed
 ge systems.\n\n \n\nNote: Please enter your institutional/corporate email
  when registering.\n\n \n\nhttps://events.grnet.gr/event/219/
LOCATION:
URL:https://events.grnet.gr/event/219/
END:VEVENT
END:VCALENDAR