CCCCatalyzing the computing research community and enabling the pursuit of innovative, high-impact research.
  • Twitter
  • Facebook
  • Youtube
  • Rss
  • About
    • About CCC
    • Council Members
    • Council Meetings
    • CCC Council Nominations
    • Governing Documents
    • FAQ
    • Contact
  • Visioning
    • Visioning Activities
      • 2022
      • 2021
      • 2020
      • 2019
      • 2018
      • 2017
      • 2016
      • 2015
      • 2014
      • 2013
      • 2012
      • 2011 and Prior Years
    • Workshop Reports
    • RFP – Creating Visions for Computing Research
    • Blue Sky
    • CS for Social Good White Paper Competition
    • Robotics Roadmap
  • Leadership Development
    • Call for Council Nominations
    • Leadership in Science Policy Institute
    • Big Data Regional Hubs
    • Postdoc Best Practices
      • Postdoc Best Practice Final Reports
      • Postdoc Best Practice Resources
    • CIFellows
      • CIFellows 2021
      • CIFellows 2020
      • CIFellows 2020: For the Record
      • CI Fellows 2014 Workshop
      • 2011 Class
      • 2010 Class
      • 2009 Class
      • Assessment
      • Diversity
      • Success Stories
  • Task Forces
    • Computing Challenges to Humanity: Climate
    • Research Ecosystem Working Group
    • NextGen AI
    • Unique Ways to Compute
    • Socio Technical Resilience
    • Computational Challenges in Healthcare
    • Past Task Forces
      • AI Working Group
      • Weird Ways to Compute
      • Security, Integrity, and Trust
      • Future of Life in a Hybrid World
      • Computing Challenges to Humanity
  • Resources
    • CCC Call for Content
    • Workshop Reports
    • CCC-Led White Papers
    • Presentations
    • CCC Responds to the Community
    • Recent CCC Activities
    • Ongoing CCC Activities
    • CIFellows Spotlight
    • Great Innovative Ideas
    • Event Videos
    • Catalyzing Computing Podcast
    • Computing Research in Action
    • Computing Research Highlights
  • Events
    • Upcoming Events
    • Special Events
    • Past Events
    • CCC at AAAS
      • CCC at AAAS 2023
      • CCC at AAAS 2022
      • CCC at AAAS 2020
      • CCC at AAAS 2019
      • CCC at AAAS 2018
      • CCC at AAAS 2017
      • CCC at AAAS 2016
      • CCC at AAAS 2013
  • CCC by CS Area
    • AI /ML / Robotics
    • Architecture / Systems / Networking
    • Databases / Informatics / Data Science / HPC
    • Human-Computer Interaction / Graphics / Visualization
    • IoT / Ubiquitous
    • Programming Languages / Compilers / Software Engineering
    • Security / Privacy / Fairness
    • Theory / Algorithms
    • Miscellaneous
  • Blog
  • Podcast
  • Search
  • Menu

Fair Representations and Fair Interactive Learning


   Workshop Report   

March 18-19, 2018

The Westin Philadelphia
The Westin Philadelphia, Philadelphia, PA, United States



Event Contact

Ann Drobnis
adrobnis@cra.org


Event Type

2018 Events, 2018 Visioning Activities, Visioning Activities, Workshop


Event Category

CCC

Overview

“Fairness” and “Machine Learning” both mean many things, which naturally complicates the study of fairness in machine learning. Hence, it is important to specialize. In this workshop, we will focus on two fair learning paradigms: “interactive” methods, and representation learning. This workshop aims to identify key challenges and open questions that currently limit both our theoretical understanding of these methods, and their applicability in practice.

Interactive machine learning methods operate in settings in which the actions of the learning algorithm feedback into its data collection, or change the environment in which it is operating, or both. This class of settings includes both bandit optimization and reinforcement learning (among others), and is distinct from batch classification.

Interactive learning settings naturally arise in many fairness-relevant learning tasks: Predictive policing algorithms are able to observe crime to higher fidelity in areas to which they have deployed officers (and may also reduce crime in areas to which more police officers have been deployed). Lending and criminal recidivism prediction algorithms, when actually used to make decisions, observe outcomes corresponding to only one of their decisions (i.e. loan payback only if the loan was granted, recidivism only if the individual was released). The feedback loops that can result when this structure has not been properly taken into account has been blamed as one source of unfairness. However, “exploration”, which is necessary to avoid these feedback loops, has also been identified as a potentially unfair operation, when the decisions of the algorithm correspond to important actions affecting the lives of individuals. How should we think about managing these tradeoffs when fairness concerns predominate?

Fair representation learning is the process of constructing transformations of the original data that retain as much of the task-relevant information as possible while removing information about sensitive or protected attributes. Informally, these may be thought of as a form of data “de-biasing”. This shifts the fairness burden away from the particular choice of prediction method towards identifying data transformations that can be used in combination with any desired modeling approach. For instance, by removing all information that can be used to infer a defendant’s race from a set of criminal data, one can construct recidivism prediction models that are in a sense “race-neutral”. Yet questions remain about what kinds of fairness such representations can effectively promote, and whether we risk introducing any other types of biases along the way. How fair are fair representations, really?

Agenda

March 18, 2018 (Sunday)

10:00 AM SNACK AVAILABLE | The Georgian Room
10:15 AM Opening Remarks | The Georgian Room
10:30 AM Introductions | The Georgian Room
11:30 AM Short Talks and Discussion: Fairness in Interactive Learning | The Georgian Room

Michael Kearns, University of Pennsylvania

Carlos Scheidegger, University of Arizona

Jamie Morgenstern, Georgia Tech

12:45 PM LUNCH | Independence Room
01:30 PM Short Talks and Discussion: Fair Representations | The Georgian Room

Christina Ilvento, Harvard University

Adam Kalai, Microsoft Research

Sorelle Friedler, Haverford College

Rich Zemel, Toronto

02:45 PM BREAK | The Georgian Room
03:00 PM Panel: Practice and Policy | The Georgian Room

Richard Berk, University of Pennsylvania

Solon Barocas, Cornell University

Kristian Lum, HRDAG

Andrew Selbst, Data & Society Research Institute

Jon Kleinberg, Cornell University

04:00 PM BREAK | The Georgian Room
04:15 PM Breakout Discussion | The Directors Room
05:15 PM Discussion Group Summaries | The Georgian Room
05:30 PM Brief Closing Remarks for Day 1 | The Georgian Room
06:30 PM DINNER | The Independence Room

March 19, 2018 (Monday)

07:30 AM BREAKFAST | The Independence Room
08:30 AM Recap and New Parallel Breakout Assignments | The Georgian Room
09:30 AM Breakout Discussion | The Directors Room
10:30 AM BREAK | The Georgian Room
11:00 AM Report Writing | The Georgian Room
12:30 PM LUNCH, Concluding Remarks | The Independence Room
Organizers

Alexandra Chouldechova (Carnegie Mellon)
Aaron Roth (University of Pennsylvania)

Logistics

The Computing Community Consortium (CCC) will cover travel expenses for all participants who desire it. Participants are asked to make their own travel arrangements to get to the workshop, including purchasing airline tickets. Following the workshop, CCC will circulate a reimbursement form that participants will need to complete and submit, along with copies of receipts for amounts exceeding $75.

In general, standard Federal travel policies apply: CCC will reimburse for non-refundable economy airfare on U.S. Flag carriers; and no alcohol will be covered.

For more information, please see the Guidelines for Participant Reimbursements from CCC.

Additional questions about the reimbursement policy should be directed to Ann Drobnis, CCC Director (adrobnis [at] cra.org).

CRA - Uniting Industry, Academia and Government to Advance Computing Research and Change the World.
CCC - Catalyzing the computing research community and enabling the pursuit of innovative, high-impact research.
Increasing the Success and Participation of Underrepresented Groups in Computing Research.
CRA-E - Addressing society’s need for a continuous supply of talented and well-educated computing researchers.
CERP - Promoting diversity in computing through evaluation and research.
Increasing interaction between industry partners and other organizations involved in computing research for the benefit of all.
CRA Home | Contact Us | Unsubscribe/Removal of Information | Terms of Use         © Copyright 2021 - CRA
Digital Computing Beyond Moore’s Law Sociotechnical Interventions for Health Disparity Reduction: A Research Age...
Scroll to top