Learning with Complex Loss Functions and Constraints

By Harikrishna Narasimhan. 2018

93 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Wasserstein Fair Classification (2019, methodology): “Some methods transform the constrained optimization problem via the method of Lagrange multipliers (Agarwal et al., 2018; Corbett-Davies et al., 2017; Cotter et al., 2018; Goh et al., 2016; Narasimhan, 2018; Wu et al., 2018; Zafar et al., 2017).”
Optimization with Non-Differentiable Constraints with Applications to Fairness, Recall, Churn, and Other Goals (2018, methodology): “Given that, what is the best way to convert a stochastic classifier into a deterministic one? Recently, Narasimhan et al. (2019b) investigate this question theoretically and experimentally.”
Lagrangian Duality for Constrained Deep Learning (2020, background): “A further collection of works interpret constrained optimization as a twoplayer game, in which one player optimizes the objective function and a second player attempt at satisfying the problem constraints [24, 32, 1].”
Fairness without Harm: Decoupled Classifiers with Preference Guarantees (2019, methodology): “Many of the proposed methods for fair machine learning have aimed to build models that predict or perform in the same way across groups (e.g., Hardt et al., 2016; Zafar et al., 2017a; Feldman et al., 2015; Zafar et al., 2017c; Agarwal et al., 2018; Narasimhan, 2018).”
User-oriented Fairness in Recommendation (2021, background): “[12, 31, 40, 44] and ranking tasks [8, 38, 39, 43], etc.”
Two-Player Games for Efficient Non-Convex Constrained Optimization (2018, background): “(2012) surveys some such work, and there are several more recent examples (e.g. Kearns et al., 2017; Narasimhan, 2018; Agarwal et al., 2018).”
Training Well-Generalizing Classifiers for Fairness Metrics and Other Data-Dependent Constraints (2018, methodology): “To this end, we build on a long line of prior work that treats constrained optimization as a two-player game [e.g. Christiano et al., 2011, Arora et al., 2012, Rakhlin and Sridharan, 2013, Kearns et al., 2017, Narasimhan, 2018, Agarwal et al., 2018].”
Fairness in Machine Learning (2020, background): “…185] or add penalties to the objective [5, 14, 22, 44, 46, 56, 58, 62, 73, 74, 75, 79, 82, 87, 88, 89, 93, 94, 95, 98, 104, 115, 119, 123, 133, 134, 135, 138, 139, 154, 165, 176, 179, 180, 182], others use adversary techniques to maximize the system ability to predict the target while minimizing…”
Fairness in Machine Learning: A Survey (2020, background): “[5, 57, 65, 124, 195, 209, 210, 288, 112, 208, 75, 284, 164]) have similar objectives to fairness regularization approaches, and hence we present them together.”
Identifying and Correcting Label Bias in Machine Learning (2019, background): “Related approaches include Edwards and Storkey (2015); Corbett-Davies et al. (2017); Narasimhan (2018).”
On the Compatibility of Privacy and Fairness (2019, methodology): “These measures of fairness belong to a broader class of fairness constraints that can be represented as constraints on functions of the confusion matrix of a classifier [18].”
Pairwise Fairness for Ranking and Regression (2019, background): “Fair Classiﬁcation Many statistical fairness metrics for binary classiﬁcation can be written in terms of rate constraints , that is, constraints on the classiﬁer’s positive (or negative) prediction rate for different groups [Goh et al., 2016, Narasimhan, 2018, Cotter et al., 2019a,b].”
Frank-Wolfe Method is Automatically Adaptive to Error Bound Condition (2018, methodology): “In recent years, the FW method has gained an increasing interest in large-scale optimization and machine learning (e.g., (Garber and Hazan, 2015; Freund and Grigas, 2016; Nesterov, 2018; Narasimhan, 2018)).”