BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//PYVOBJECT//NONSGML Version 1//EN
BEGIN:VTIMEZONE
TZID:Europe/Brussels
BEGIN:STANDARD
DTSTART:20001029T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CET
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T020000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:20261007T190346Z - 21853@eu441a.odoo.com
DTSTART;TZID=Europe/Brussels:20261123T090000
DTEND;TZID=Europe/Brussels:20261125T130000
CREATED:20261007T190346Z
DESCRIPTION:<a href="https://www.sustain.brussels/event/training-track-from
 -data-to-prediction-276/register">Training Track: From Data to Prediction 
  </a>\nCourse Description This course gives non-technical decision-makers 
 the skills to evaluate AI and machine learning tools. Participants learn t
 o challenge vendor claims\, spot bias\, and judge whether a prediction or 
 solution can be trusted\, using simple frameworks and the right questions 
 to cut through the hype. Course Content Day Content 1 Thinking Tools for P
 rediction: Critical thinking foundations. Basic AI and machine learning li
 teracy. The science behind AI tools. 2 The evaluation grid\, part 1 : Key 
 concepts and case studies: consistency\, robustness\, quality of abstentio
 n\, predictability\, accuracy\, confidence\, safety\, compliance\, severit
 y of harm\, human override\, transparency 3 The evaluation grid\, part 2\,
  and current limits of AI: Performance metrics\, data quality\, overfittin
 g and data leakage\, gaps between what a model predicts and what you actua
 lly want\, reproducibility. Explainability and interpretability\, bias and
  fairness\, dependency\, accountability\, privacy. Learning Outcomes By th
 e end of the course\, participants can:Question vendor claims and read mod
 el documentation with a critical eyeSpot bias\, overfitting\, and other wa
 rning signs in AI solutionsUse grids and checklists to judge a solution's 
 reliability and fairnessTell real capability from hype\, and match the rig
 ht type of AI to the jobExplain the risks of a model in plain language to 
 colleagues and decision-makersPractical Work Thought experimentsEvaluation
  grid walkthroughAudit case study\; evaluation memo draftingDeliverables M
 L Solution Evaluation Grid — structured assessment grid across accuracy\
 , fairness\, transparency\, etc.Red Flags Checklist — reusable tool for 
 evaluating ML/AI proposalModel Evaluation Memo — one-page non-technical 
 recommendation for senior leadership Target Audience: Non-technical profes
 sionals from public or private organisations\, including managers\, procur
 ement officers\, [...]
DTSTAMP:20261007T190346Z
LOCATION:BeCentral\, Cantersteen 12\, 1000 Bruxelles\, Belgium
SUMMARY:Training Track: From Data to Prediction  
X-ALT-DESC;FMTTYPE=text/html:<a href="https://www.sustain.brussels/event/tr
 aining-track-from-data-to-prediction-276/register">Training Track: From Da
 ta to Prediction  </a>\nCourse Description This course gives non-technical
  decision-makers the skills to evaluate AI and machine learning tools. Par
 ticipants learn to challenge vendor claims\, spot bias\, and judge whether
  a prediction or solution can be trusted\, using simple frameworks and the
  right questions to cut through the hype. Course Content Day Content 1 Thi
 nking Tools for Prediction: Critical thinking foundations. Basic AI and ma
 chine learning literacy. The science behind AI tools. 2 The evaluation gri
 d\, part 1 : Key concepts and case studies: consistency\, robustness\, qua
 lity of abstention\, predictability\, accuracy\, confidence\, safety\, com
 pliance\, severity of harm\, human override\, transparency 3 The evaluatio
 n grid\, part 2\, and current limits of AI: Performance metrics\, data qua
 lity\, overfitting and data leakage\, gaps between what a model predicts a
 nd what you actually want\, reproducibility. Explainability and interpreta
 bility\, bias and fairness\, dependency\, accountability\, privacy. Learni
 ng Outcomes By the end of the course\, participants can:Question vendor cl
 aims and read model documentation with a critical eyeSpot bias\, overfitti
 ng\, and other warning signs in AI solutionsUse grids and checklists to ju
 dge a solution's reliability and fairnessTell real capability from hype\, 
 and match the right type of AI to the jobExplain the risks of a model in p
 lain language to colleagues and decision-makersPractical Work Thought expe
 rimentsEvaluation grid walkthroughAudit case study\; evaluation memo draft
 ingDeliverables ML Solution Evaluation Grid — structured assessment grid
  across accuracy\, fairness\, transparency\, etc.Red Flags Checklist — r
 eusable tool for evaluating ML/AI proposalModel Evaluation Memo — one-pa
 ge non-technical recommendation for senior leadership Target Audience: Non
 -technical professionals from public or private organisations\, including 
 managers\, procurement officers\, [...]
END:VEVENT
END:VCALENDAR