Time in GMT (UTC+00). Please check for your corresponding local time on timeanddate.com
Full Schedule with links to zoom: https://neurips.cc/virtual/2021/workshop/21850
Information Booklet : https://drive.google.com/file/d/1Bbq2DAnScbUufcwvxwkqCtF1lvc5TD1s/view
Invited Talk
09:10 AM - 09:34 AM GMT
Rumi Chunara
Abstract: By leveraging principles of health equity, I will discuss the use of causal models and machine learning to address realistic challenges of data collection and model use across environments. Examples include a domain adaptation approach that improves prediction in under-represented population sub-groups by leveraging invariant information across groups when possible, and an algorithmic fairness method which specifically incorporates structural factors to better account for and address sources of bias and disparities.
Q&A
09:34 AM - 09:45 AM GMT
Short break: Join us on Gathertown
Invited Talk
09:50 AM - 10:20 AM GMT
Silvia Chiappa
Abstract: Arguably, the consideration of causal effects along subsets of causal paths is required for understanding and addressing unfairness in most real-world scenarios. In this talk I will share some of my thoughts on this goal and on the challenges to achieve it.
Q&A
10:20 AM - 10:30 AM GMT
Short break: Join us on Gathertown
Invited Talk
10:40 AM - 11:10 AM GMT
Isabel Valera
Abstract: In the recent years, we have observed an explosion of research approaches at the intersection of causality and fairness in machine learning (ML). These works are often motivated by the promise that causality allows us to reason about the causes of unfairness both in the data and in the ML algorithm. However, the promises of existing causal fair approaches require strong assumptions, which hinder their practical application. In this talk, I will provide a quick overview of both the promises and the technical challenges of causal fair ML frameworks from a theoretical perspective. Finally, I will show how to leverage probabilistic ML to partially relax causal assumptions in order to develop more practical solutions to causal fair ML.
Q&A
11:10 AM - 11:20 AM GMT
Contributed Talks
11:20 AM - 11:30 AM GMT
Agnieszka Słowik · Leon Bottou
11:30 AM - 11:40 AM GMT
Nikola Konstantinov · Christoph Lampert
11:40 AM - 11:50 AM GMT
Wen Huang · Lu Zhang · Xintao Wu
Q&A
11:50 AM - 12:00 PM GMT
12:00 PM - 01:00 PM GMT
Discussions
1:00 PM - 2:00 PM GMT
Leads: Issa Kohler-Hausmann, Matt Kusner, Maggie Makar, Ioana Bica
Leads: Silvia Chiappa, Alex D’Amour, Elliot Creager
Leads: Isabel Valera, Ulrich Aïvodji, Keziah Naggita, Stephen Pfohl
Leads: Luke Stark, Irene Y. Chen, Lizzie Kumar
Long break: join us on Gathertown
Invited Talk
4:00 PM - 4:30 PM GMT
Elias Bareinbiom
Q&A
4:30 PM - 4:40 PM GMT
Invited Talk
4:40 PM - 5:20 PM GMT
Hima Lakkaraju
Abstract: As machine learning black boxes are increasingly being deployed in domains such as healthcare and criminal justice, there is growing emphasis on building tools and techniques for explaining these black boxes in an interpretable manner. Such explanations are being leveraged by domain experts to diagnose systematic errors and underlying biases of black boxes. In this talk, I will present some of our recent research that sheds light on the vulnerabilities of popular post hoc explanation techniques such as LIME and SHAP, and also introduce novel methods to address some of these vulnerabilities. More specifically, I will first demonstrate that these methods are brittle, unstable, and are vulnerable to a variety of adversarial attacks. Then, I will discuss two solutions to address some of the aforementioned vulnerabilities–(i) a Bayesian framework that captures the uncertainty associated with post hoc explanations and in turn allows us to generate explanations with user specified levels of confidence, and (ii) a framework based on adversarial training that is designed to make post hoc explanations more stable and robust to shifts in the underlying data; I will conclude the talk by discussing our recent theoretical results which shed light on the equivalence and robustness of state-of-the-art explanation methods.
Q&A
5:20 PM - 5:30 PM GMT
Contributed Talks
5:30 PM - 5:40 PM GMT
Irene Y Chen · Hal Daumé III · Solon Barocas
5:40 PM - 5:50 PM GMT
Subhabrata Majumdar · Cheryl Flynn · Ritwik Mitra
5:50 PM - 5:52 PM GMT
Anshuman Chhabra · Adish Singla · Prasant Mohapatra
Q&A
5:52 PM - 6:05 PM GMT
Short break
Invited Talk
6:15 PM - 6:47 PM GMT
Aditi Raghunathan
Abstract: Current machine learning (ML) methods are primarily centered around improving in-distribution generalization where models are evaluated on new points drawn from nearly the same distribution as the training data. On the other hand, robustness and fairness involve reasoning about out-of-distribution performance such as accuracy on protected groups or perturbed inputs, and reliability even in the presence of spurious correlations. In this talk, I will describe an important lesson from robustness: in order to improve out-of-distribution performance, we often need to question the common assumptions in ML. In particular, we will see that ‘more data’, ‘bigger models’, or ‘fine-tuning pretrained features’ which improve in-distribution generalization often fail out-of-distribution.
Q&A
6:47 PM - 6:55 PM GMT
Short break
Discussions
7:00 PM - 7:40 PM GMT
7:40 PM - 08:20 PM GMT
