4:00 PM - 6:15 PM June 22, 2021 KST
3:00 AM - 5:15 AM June 22, 2021 EDT
0:00 AM - 3:10 PM June 22, 2021 PDT
The target audience is anyone who is interested in machine learning, especially data scientists and practitioners who want to learn about shortcut learning bias. The tutorial assumes some basic knowledge of machine learning tasks.
In this tutorial, we will outline notions of shortcut learning and discuss methods for alleviating the issue. Specifically,
We will exemplify the shortcut learning bias by demonstrating the challenges that modern deep learning models encounter.
We will formalize the notion of shortcut learning bias through the lens of causality and invariance.
We will review the state-of-the-art techniques for addressing bias.
We hope our tutorial will alert the shortcut learning bias risk and encourage developing fair, accountable, and transparent ML methods for experts in the FAccT community.
Part 1. Introduction to shortcut learning (15min) [Link]
Part 2. Understanding Shortcut Learning through the Lens of Causality & Invariance (30min) [Link] [Tech Report]
Part 3. Learning Invariant Representations (30min) [Link]
Part 4. How to mitigate Shortcut Learning in practice? & Future research directions (15min) [Link]
A tech report corresponding to Part 2. Understanding Shortcut Learning through the Lens of Causality & Invariance (30min) is here: