BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//pyconde-pydata-2026//speaker//QVC73Q
BEGIN:VTIMEZONE
TZID:CET
BEGIN:STANDARD
DTSTART:20001029T040000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CET
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-pyconde-pydata-2026-BAXEXY@pretalx.com
DTSTART;TZID=CET:20260414T171000
DTEND;TZID=CET:20260414T174000
DESCRIPTION:### Agent-Based Hyperparameter Optimization for Gradient Booste
 d Trees\n\nHyperparameter optimization for gradient boosted tree models is
  a repetitive yet cognitively demanding task. Practitioners must combine s
 tatistical intuition with detailed\, library-specific knowledge—often bu
 ried across hundreds of pages of documentation for tools such as XGBoost\,
  LightGBM\, or CatBoost. As models and configurations grow in complexity\,
  traditional approaches like grid search\, random search\, or even Bayesia
 n optimization struggle to incorporate semantic understanding of model beh
 avior.\n\nUsing XGBoost as a concrete case study\, I demonstrate how RAG-p
 owered agents\, orchestrated in a structured workflow\, can analyze model 
 behavior via SHAP values\, diagnose failure modes (e.g. overfitting\, feat
 ure dominance\, interaction leakage)\, and propose targeted hyperparameter
  adjustments grounded in both theory and library-specific constraints.\n\n
 The system combines open-source tools including XGBoost\, SHAP\, Optuna\, 
 and LangGraph/LangChain\, where agents specialize in tasks such as model d
 iagnostics\, documentation-aware parameter reasoning\, and experiment orch
 estration. Rather than replacing existing optimization frameworks\, agents
  operate on top of them\, injecting domain knowledge and interpretability 
 signals into the optimization loop.
DTSTAMP:20260412T141727Z
LOCATION:Ferrum [2nd Floor]
SUMMARY:Agent-Based Hyperparameter Optimization for Gradient Boosted Trees 
 - Huijo Kim
URL:https://pretalx.com/pyconde-pydata-2026/talk/BAXEXY/
END:VEVENT
END:VCALENDAR