BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Drupal iCal API//EN
X-WR-CALNAME:Events items teaser
X-WR-TIMEZONE:America/Toronto
BEGIN:VTIMEZONE
TZID:America/Toronto
X-LIC-LOCATION:America/Toronto
BEGIN:DAYLIGHT
TZNAME:EDT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
DTSTART:20240310T070000
END:DAYLIGHT
BEGIN:STANDARD
TZNAME:EST
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
DTSTART:20241103T060000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
UID:69d1ea2216e05
DTSTART;TZID=America/Toronto:20250214T153000
SEQUENCE:0
TRANSP:TRANSPARENT
DTEND;TZID=America/Toronto:20250214T163000
URL:https://uwaterloo.ca/combinatorics-and-optimization/events/tutte-colloq
 uium-xi-he
SUMMARY:Tutte colloquium-Xi He
CLASS:PUBLIC
DESCRIPTION:TITLE:Accuracy Aware Minimally Invasive Data Exploration For De
 cision\nSupport\n\nSPEAKER:\n Xi He\n\nAFFILIATION:\n University of Waterl
 oo\n\nLOCATION:\n MC 5501\n\nABSTRACT: Decision-support (DS) applications
 \, crucial for timely and\ninformed decision-making\, often analyze sensit
 ive data\, raising\nsignificant privacy concerns. While privacy-preserving
  randomized\nmechanisms can mitigate these concerns\, they introduce the r
 isk of\nboth false positives and false negatives. Critically\, in DS\nappl
 ications\, the number of false negatives often needs to be strictly\ncontr
 olled. Existing privacy-preserving techniques like differential\nprivacy\,
  even when adapted\, struggle to meet this requirement without\nsubstantia
 l privacy leakage\, particularly when data distributions are\nskewed. This
  talk introduces a novel approach to minimally invasive\ndata exploration 
 for decision support. Our method minimizes privacy\nloss while guaranteein
 g a bound on false negatives by dynamically\nadapting privacy levels based
  on the underlying data distribution. We\nfurther extend this approach to 
 handle complex DS queries\, which may\ninvolve multiple conditions on dive
 rse aggregate statistics combined\nthrough logical disjunction and conjunc
 tion. Specifically\, we define\ncomplex DS queries and their associated ac
 curacy requirements\, and\npresent algorithms that strategically allocate 
 a privacy budget to\nminimize overall privacy loss while satisfying the bo
 unded accuracy\nguarantee.\n\n \n\n 
DTSTAMP:20260405T045042Z
END:VEVENT
END:VCALENDAR