Responsive image

David Madras

PhD Student

I'm a PhD Student in the Machine Learning Group at the University of Toronto and the Vector Institute. My supervisor is Rich Zemel. I'm primarily interested in how to learn better and fairer algorithmic decision-making systems. My interests include fairness, causal inference, and generative modelling.

Announcements

  • June 18, 2020 — ICML Acceptance

    Our paper Causal Modelling for Fairness in Dynamical Systems (with Elliot Creager, Toni Pitassi, and Rich Zemel) was accepted to ICML 2020!

  • December 19, 2019 — ICLR Acceptance

    Our paper Detecting Extrapolation with Local Ensembles (with James Atwood and Alex D'Amour) was accepted to ICLR 2020!

  • November 23-24, 2019 — Pan-Canadian Self-Organizing Conference on Machine Learning

    Excited to have co-organized and program chair the inaugural Pan-Canadian Self-Organizing Conference on Machine Learning (PC-SOCMLx)! This event is for Canadian graduate students in machine learning to meet each other and develop a research community. Thanks to the Vector Institute, Mila, Amii, CIFAR, and Facebook for their support in making this event happen. The feedback from the event was great and we're looking forward to the next one!

  • September 26-28, 2019 — NBER Economics of Artificial Intelligence Conference

    Excited to be attending the NBER Economics of Artificial Intelligence Conference, hosted at the University of Toronto, along with the NBER Economics of AI Young Scholars Workshop beforehand. Looking forward to meeting lots of interesting people from far outside my research area!

  • September 23, 2019— New Preprint

    Causal Modeling for Fairness in Dynamical Systems is on arXiv! We discuss how to model standard problems of long-term unfairness in systems using causal graphical models, and demonstrate the advantages of this approach through detailed case studies.

  • July 21-24, 2019 — Edmonton - The Summer Institute on AI and Society

    Excited to be attending The Summer Institute on AI and Society, jointly hosted by CIFAR, AMII, and UCLA Law! Looking forward to meeting interesting people from across a range of disciplines.

  • May 19, 2019 — ICML Workshop Acceptance

    Our paper "Detecting Extrapolation with Influence Functions" (with James Atwood and Alex D'Amour) was accepted as a contributed talk to the Workshop on Uncertainty and Robustness in Deep Learning at ICML 2019!

  • May 12, 2019 — Kigali - African Institute for Mathematical Sciences

    I'm very excited to be in Kigali, Rwanda until the end of May, to teach a course on Privacy and Fairness in Machine Learning at the African Master's in Machine Intelligence, along with Elliot Creager, Toni Pitassi, and Rich Zemel.

  • April 22, 2019 — ICML Acceptance

    Our paper "Flexibly Fair Representation Learning by Disentanglement" (with Elliot Creager, Jorn Jacobsen, Marissa Weis, Kevin Swersky, Toni Pitassi, and Rich Zemel) was accepted to ICML 2019!

  • February 16, 2019 — Cambridge - Google Brain

    I'll be working as a research intern at Google Brain in Cambridge until mid-May, under the supervision of Alex D'Amour, thinking about causal inference and out-of-distribution detection.

  • February 5, 2019 — Princeton

    In Princeton this week! I'm giving a talk at 12:20pm in the CS building, room 201 on Tuesday.

  • November 18, 2018 — FAT* Acceptance

    Fairness Through Causal Awareness: Learning Causal Latent-Variable Models for Biased Data was accepted to FAT* 2019! See the list of accepted papers here. I'll also be presenting it at the Workshop on Ethical, Social, and Governance Issues in AI in December.

  • September 14, 2018 — Attended Fairness in ML Workshop

    I attended the ML Fairness Workshop at Google in Cambridge, MA, USA. I was excited to meet a number of very interesting people doing research in fair ML - thanks to Google for hosting!

  • September 7, 2018 — New Preprint

    Fairness Through Causal Awareness: Learning Causal Latent-Variable Models for Biased Data is on arXiv! We propose using causal inference and generative modeling to better learn from historically biased datasets.

  • September 7, 2018 — Code on Github

    Our code for Learning Adversarially Fair and Transferable Representations is on Github - thanks to my collaborator Elliot Creager for all his hard work on this code.

  • September 4, 2018 — NIPS Acceptance

    Predict Responsibly: Improving Fairness and Accuracy by Learning to Defer was accepted to NIPS 2018!

  • Papers

    Publications

  • Causal Modeling for Fairness in Dynamical Systems
    Elliot Creager, David Madras, Toniann Pitassi, Richard Zemel
    International Conference on Machine Learning, 2020.
  • Detecting Extrapolation with Local Ensembles
    David Madras, James Atwood, Alex D'Amour
    International Conference on Learning Representations, 2020.
  • Flexibly Fair Representation Learning by Disentanglement
    Elliot Creager, David Madras, Joen-Henrik Jacobsen, Marissa Weis, Kevin Swersky, Toniann Pitassi, Richard Zemel
    International Conference on Machine Learning, 2019.
  • Fairness Through Causal Awareness: Learning Causal Latent-Variable Models for Biased Data
    David Madras, Elliot Creager, Toniann Pitassi, Richard Zemel
    ACM Conference on Fairness, Accountability, and Transparency (ACM FAT*) 2019.
  • Predict Responsibly: Improving Fairness and Accuracy by Learning to Defer
    David Madras, Toniann Pitassi, Richard Zemel
    Neural Information Processing Systems, 2018.
  • Learning Adversarially Fair and Transferable Representations
    David Madras*, Elliot Creager*, Toniann Pitassi, Richard Zemel
    International Conference on Machine Learning, 2018 (Oral).
  • Improving fairness in match play golf through enhanced handicap allocation
    Timothy C. Y. Chan, David Madras, Martin Puterman
    Journal of Sports Analytics, vol. 4, no. 4, pp. 251-262, 2018.
  • Change-point Detection Methods for Body-Worn Video
    Stephanie Allen*, David Madras*, Ye Ye*, Greg Zanotti*
    SIURO (Vol. 10), August 2017 and Joint Math Meetings (JMM) 2017.
  • Slides

    Learning Adversarially Fair and Transferable Representations: ICML 2018 (Oral)

    Fairness in Machine Learning: Princess Margaret Hospital Summer Series 2018

    Fairness in Machine Learning: An Overview: U of T Undergraduate AI Group (UAIG) AI Day, 2017

    Stochastic Variational Inference: Tutorial for CSC412/2506: Probabilistic Learning and Reasong

    Unifying Visual-Semantic Embeddings with Multimodal Neural Language Models: paper presentation for CSC2539, Topics in Computer Vision: Visual Recognition with Text

    Differentially Private Recommender Systems: paper presentation for CSC2419, Topics in Cryptography: Algorithms & Complexity in Private Data Analysis

    Other

    I'm also a musician - I love writing, singing, and playing music. I put out an album in August 2018 - if you're interested, check it out on Spotify or see my my website for more info! A couple of years ago I wrote the songs for a musical in the Toronto Fringe Festival. I also play jazz piano and love to improvise.

    Contact

    Email: lastname (at) cs (dot) toronto (dot) edu