An Axiomatic Theory of Provably-Fair Welfare-Centric Machine Learning
Dec 6, 2021
Speakers
Cyrus Cousins
Speaker · 0 followers
About
We address an inherent difficulty in welfare-theoretic fair ML, propose an alternative, and study the resulting computational and statistical learning questions. Welfare metrics quantify overall wellbeing across a population of groups, and welfare-based objectives and constraints have recently been proposed to incentivize fair ML methods to satisfy their diverse needs. However, many ML problems are cast as loss minimization, rather than utility maximization tasks, thus requiring non-trivial modeling to construct utility functions. We define a complementary metric, termed malfare, measuring overall societal harm, with axiomatic justification via the standard axioms of cardinal welfare, and cast fair ML as malfare minimization over the risk values (expected losses) of each group. Surprisingly, the axioms of cardinal welfare (malfare) dictate that this is not equivalent to simply defining utility as negative loss and maximizing welfare. Building upon these concepts, we define fair-PAC learning, where a fair PAC-learner is an algorithm that learns an ε-δ malfare-optimal model, with bounded sample complexity, for any data distribution and (axiomatically justified) malfare concept. We show conditions under which many standard PAC-learners may be converted to fair-PAC learners. This places fair-PAC learning on firm theoretical ground, as it yields statistical, and in some cases computational, efficiency guarantees for many well-studied machine-learning models, and is also practically relevant, as it democratizes fair ML by providing concrete training algorithms and rigorous generalization guarantees.We address an inherent difficulty in welfare-theoretic fair ML, propose an alternative, and study the resulting computational and statistical learning questions. Welfare metrics quantify overall wellbeing across a population of groups, and welfare-based objectives and constraints have recently been proposed to incentivize fair ML methods to satisfy their diverse needs. However, many ML problems are cast as loss minimization, rather than utility maximization tasks, thus requiring non-trivial mode…
Organizer
NeurIPS 2021
Account · 1.9k followers
About NeurIPS 2021
Neural Information Processing Systems (NeurIPS) is a multi-track machine learning and computational neuroscience conference that includes invited talks, demonstrations, symposia and oral and poster presentations of refereed papers. Following the conference, there are workshops which provide a less formal setting.
Like the format? Trust SlidesLive to capture your next event!
Professional recording and live streaming, delivered globally.
Sharing
Recommended Videos
Presentations on similar topic, category or speaker
Beyond Bandit Feedback in Online Multiclass Classification
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
SNIPS: Solving Noisy Inverse Problems Stochastically
Watch later
Bahjat Kawar, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Off-Policy Correction For Multi-Agent Reinforcement Learning
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Revenue maximization via machine learning with noisy data
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Implicit Semantic Response Alignment for Partial Domain Adaptation
Watch later
Wenxiao Xiao, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Predicting Power System Dynamics and Transients: A Frequency Domain Approach
Watch later
Wenqi Cui, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%