Difference between revisions of "Fairness Without Demographics in Repeated Loss Minimization"

From statwiki
Jump to: navigation, search
(Risk Bounding Over Unknown Groups)
(Risk Bounding Over Unknown Groups)
Line 15: Line 15:
 
==Risk Bounding Over Unknown Groups==
 
==Risk Bounding Over Unknown Groups==
  
At this point our goal is to minimize the worst-case group risk over a single time-step <math display="inline">\mathcal{R}_{max} (\theta^{(t)}) </math>. As previously mentioned, this is difficult to do because neither the population proportions <math display="inline">\{a_k\} </math> nor group distributions <math display="inline">\{P_k\} </math> are known. Therefore, Hashimoto et al. developed an optimization technique that is robust "against '''''all''''' directions around the data generating distribution". This refers to the notion that DRO is robust to any group distribution <math display="inline">P_k </math> whose loss other optimization techniques such as ERM might try to optimize. To create this distributionally robustness, the optimizations risk function <math display="inline">\mathcal{R}_{dro} </math> has to "up-weigh" data <math display="inline">Z</math> that cause high loss <math display="inline">\ell(\theta, Z)</math>. In other words, the risk function has to over-represent mixture components (i.e. group distributions <math display="inline">\{P_k\} </math>) in relation to their original mixture weights (i.e. the population proportions <math display="inline">\{a_k\} </math>) for groups that suffer high loss.  
+
At this point our goal is to minimize the worst-case group risk over a single time-step <math display="inline">\mathcal{R}_{max} (\theta^{(t)}) </math>. As previously mentioned, this is difficult to do because neither the population proportions <math display="inline">\{\alpha_k\} </math> nor group distributions <math display="inline">\{P_k\} </math> are known. Therefore, Hashimoto et al. developed an optimization technique that is robust "against '''''all''''' directions around the data generating distribution". This refers to the notion that DRO is robust to any group distribution <math display="inline">P_k </math> whose loss other optimization techniques such as ERM might try to optimize. To create this distributionally robustness, the optimizations risk function <math display="inline">\mathcal{R}_{dro} </math> has to "up-weigh" data <math display="inline">Z</math> that cause high loss <math display="inline">\ell(\theta, Z)</math>. In other words, the risk function has to over-represent mixture components (i.e. group distributions <math display="inline">\{P_k\} </math>) in relation to their original mixture weights (i.e. the population proportions <math display="inline">\{\alpha_k\} </math>) for groups that suffer high loss.  
  
 
To do this we need to consider the worst-case loss (i.e. the highest risk) over all perturbations <math display="inline">P_k </math> around <math display="inline">P</math> within a certain limit. This limit is described by the <math display="inline">\chi^2</math>-divergence (i.e. the distance, roughly speaking) between probability distributions. For two distributions <math display="inline">P</math> and <math display="inline">Q</math> the divergence is defined as <math display="inline">D_{\chi^2} (P || Q):= \int (\frac{dP}{dQ} - 1)^2</math>. With the help of the <math display="inline">\chi^2</math>-divergence, Hashimoto et al. define the chi-squared ball <math display="inline">\mathcal{B}(P,r)</math> around the probability distribution P. This ball is defined so that <math display="inline">\mathcal{B}(P,r) := \{Q \ll P : D_{\chi^2} (Q || P) \leq r \}</math>. It is exactly this ball that gives us the opportunity to consider the worst-case loss (i.e. the highest risk) over all perturbations <math display="inline">P_k </math> that lie inside the ball (i.e. within reasonable range) around the probability distribution <math display="inline">P</math>. This loss is given by
 
To do this we need to consider the worst-case loss (i.e. the highest risk) over all perturbations <math display="inline">P_k </math> around <math display="inline">P</math> within a certain limit. This limit is described by the <math display="inline">\chi^2</math>-divergence (i.e. the distance, roughly speaking) between probability distributions. For two distributions <math display="inline">P</math> and <math display="inline">Q</math> the divergence is defined as <math display="inline">D_{\chi^2} (P || Q):= \int (\frac{dP}{dQ} - 1)^2</math>. With the help of the <math display="inline">\chi^2</math>-divergence, Hashimoto et al. define the chi-squared ball <math display="inline">\mathcal{B}(P,r)</math> around the probability distribution P. This ball is defined so that <math display="inline">\mathcal{B}(P,r) := \{Q \ll P : D_{\chi^2} (Q || P) \leq r \}</math>. It is exactly this ball that gives us the opportunity to consider the worst-case loss (i.e. the highest risk) over all perturbations <math display="inline">P_k </math> that lie inside the ball (i.e. within reasonable range) around the probability distribution <math display="inline">P</math>. This loss is given by
  
 
\begin{align}
 
\begin{align}
\mathcal{R}_{dro}(\theta, r) := \underset{Q \in \mathcal{B}(P,r)}{sup} \mathbb{E}_Q [\ell(\theta;Z)].
+
\mathcal{R}_{dro}(\theta, r) := \underset{Q \in \mathcal{B}(P,r)}{sup} \mathbb{E}_Q [\ell(\theta;Z)]
 
\end{align}
 
\end{align}
 +
 +
which for <math display="inline">P:= \sum_{k \in [K]} \alpha_k P_k</math> for all models <math display="inline">\theta \in \Theta</math>  where <math display="inline">r_k := (1/a_k -1)^2</math> bounds the risk <math display="inline">\mathcal{R}_k(\theta) \leq \mathcal{R}_{dro} (\theta; r_k)</math> for each group with risk <math display="inline">\mathcal{R}_k(\theta)</math>. Furthermore, if we specify a lower bound on the group proportions <math display="inline">\alpha_{min} \leq min_{k \in [K]} \alpha_k</math>, and define <math display="inline">r_{max} := (1/\alpha_{min} -1)^2</math>, the worst-case risk <math display="inline">\mathcal{R}_{max} (\theta) </math> can be controlled by <math display="inline">\mathcal{R}_{dro} (\theta; r_{max}) </math> by forming an upper bound that can be minimized.

Revision as of 10:30, 20 October 2018

This page contains the summary of the paper "Fairness Without Demographics in Repeated Loss Minimization" by Hashimoto, T. B., Srivastava, M., Namkoong, H., & Liang, P. which was published at the International Conference of Machine Learning (ICML) in 2018. In the following, an

Overview of the Paper

Introduction

Fairness

Example and Problem Setup

Why Empirical Risk Minimization (ERM) does not work

Distributonally Robust Optimization (DRO)

Risk Bounding Over Unknown Groups

At this point our goal is to minimize the worst-case group risk over a single time-step [math]\mathcal{R}_{max} (\theta^{(t)}) [/math]. As previously mentioned, this is difficult to do because neither the population proportions [math]\{\alpha_k\} [/math] nor group distributions [math]\{P_k\} [/math] are known. Therefore, Hashimoto et al. developed an optimization technique that is robust "against all directions around the data generating distribution". This refers to the notion that DRO is robust to any group distribution [math]P_k [/math] whose loss other optimization techniques such as ERM might try to optimize. To create this distributionally robustness, the optimizations risk function [math]\mathcal{R}_{dro} [/math] has to "up-weigh" data [math]Z[/math] that cause high loss [math]\ell(\theta, Z)[/math]. In other words, the risk function has to over-represent mixture components (i.e. group distributions [math]\{P_k\} [/math]) in relation to their original mixture weights (i.e. the population proportions [math]\{\alpha_k\} [/math]) for groups that suffer high loss.

To do this we need to consider the worst-case loss (i.e. the highest risk) over all perturbations [math]P_k [/math] around [math]P[/math] within a certain limit. This limit is described by the [math]\chi^2[/math]-divergence (i.e. the distance, roughly speaking) between probability distributions. For two distributions [math]P[/math] and [math]Q[/math] the divergence is defined as [math]D_{\chi^2} (P || Q):= \int (\frac{dP}{dQ} - 1)^2[/math]. With the help of the [math]\chi^2[/math]-divergence, Hashimoto et al. define the chi-squared ball [math]\mathcal{B}(P,r)[/math] around the probability distribution P. This ball is defined so that [math]\mathcal{B}(P,r) := \{Q \ll P : D_{\chi^2} (Q || P) \leq r \}[/math]. It is exactly this ball that gives us the opportunity to consider the worst-case loss (i.e. the highest risk) over all perturbations [math]P_k [/math] that lie inside the ball (i.e. within reasonable range) around the probability distribution [math]P[/math]. This loss is given by

\begin{align} \mathcal{R}_{dro}(\theta, r) := \underset{Q \in \mathcal{B}(P,r)}{sup} \mathbb{E}_Q [\ell(\theta;Z)] \end{align}

which for [math]P:= \sum_{k \in [K]} \alpha_k P_k[/math] for all models [math]\theta \in \Theta[/math] where [math]r_k := (1/a_k -1)^2[/math] bounds the risk [math]\mathcal{R}_k(\theta) \leq \mathcal{R}_{dro} (\theta; r_k)[/math] for each group with risk [math]\mathcal{R}_k(\theta)[/math]. Furthermore, if we specify a lower bound on the group proportions [math]\alpha_{min} \leq min_{k \in [K]} \alpha_k[/math], and define [math]r_{max} := (1/\alpha_{min} -1)^2[/math], the worst-case risk [math]\mathcal{R}_{max} (\theta) [/math] can be controlled by [math]\mathcal{R}_{dro} (\theta; r_{max}) [/math] by forming an upper bound that can be minimized.