Back to All Events

EDS - Differential Privacy Demystified: How Algorithms Protect Your Data

  • NorQuest College’s Singhmar Centre for Learning 10215 108 Street Northwest Edmonton, AB, T5J 1L6 Canada (map)

We’re back with another engaging evening of data science learning and community connection! After the success of our past gatherings, we’re excited to welcome you to the next Edmonton Data Science Community event on Thursday, June 26th at NorQuest College’s Singhmar Centre for Learning. Whether you’re a privacy enthusiast, an AI professional, or just data-curious, this session is for you.

​We’re thrilled to welcome Paul Saunders, MSc student and active organizer in the local data scene, to guide us through a concept at the heart of modern data privacy: Differential Privacy.

Event Schedule:

  • Doors Open | 5:30 PM

  • Organizer's Presentation | 6:00 PM

  • Main Event (Talk by Paul Saunders) | 6:10 – 6:45 PM

  • Networking | 6:45 – 8:00 PM

Presentation Abstract:
Data privacy is a long-standing concern across numerous industries, with various methods for protecting and ensuring it. While keeping data private guarantees perfect privacy, there exist scenarios in which private data must be or should be released to the public. From censuses bound by law to ensure the privacy of citizens to machine learning models trained on healthcare data or intellectual property, strong incentives lead private information to be used within algorithms and released to the public. Hence, researchers have developed methodologies to guarantee the preservation of privacy while ensuring such data contributes to the utility and usefulness of algorithm outputs.

​One such methodology, Differential Privacy, has seen explosive growth since its introduction in the early 2000s. It's now used by multinational companies like Apple, Google, and even the United States Census Bureau. In this presentation, MSc student Paul Saunders will provide an intuitive overview of differential privacy, discussing its benefits, drawbacks, and why you may hear more about this property of algorithms in coming years.

Speaker Bio:
Paul Saunders is a current Master's Student at the University of Alberta, researching Differential Privacy and Reinforcement Learning under supervisors Dr. Nidhi Hegde and Dr. Levi Lelis. He's a past President of the Undergraduate AI Society at the University of Alberta where he completed his undergrad in 2023, a current Organizer of the Edmonton Data Society, and leads the Tea Time Talk presentation series in the RLAI lab. In his spare time, he enjoys bouldering, playing water polo, and spending time with his partner Nick.