Randomized Classifiers vs Human Decision-Makers: Trustworthy AI May Have to Act Randomly and Society Seems to Accept This

TL;DR

This paper argues that incorporating randomness into AI classifiers can enhance fairness and ethical decision-making, supported by theoretical analysis and societal acceptance evidence, challenging the preference for deterministic AI systems.

Contribution

It introduces the concept that randomized classifiers can be optimal in supervised learning, providing a theoretical foundation and empirical evidence for their societal acceptance.

Findings

Random classifiers can perform at least as well as deterministic ones.

Society shows positive attitudes towards randomized AI decision-makers.

Randomization may be necessary for achieving optimal and fair outcomes in AI decisions.

Abstract

As \emph{artificial intelligence} (AI) systems are increasingly involved in decisions affecting our lives, ensuring that automated decision-making is fair and ethical has become a top priority. Intuitively, we feel that akin to human decisions, judgments of artificial agents should necessarily be grounded in some moral principles. Yet a decision-maker (whether human or artificial) can only make truly ethical (based on any ethical theory) and fair (according to any notion of fairness) decisions if full information on all the relevant factors on which the decision is based are available at the time of decision-making. This raises two problems: (1) In settings, where we rely on AI systems that are using classifiers obtained with supervised learning, some induction/generalization is present and some relevant attributes may not be present even during learning. (2) Modeling such decisions as games reveals that any -- however ethical -- pure strategy is inevitably susceptible to exploitation. Moreover, in many games, a Nash Equilibrium can only be obtained by using mixed strategies, i.e., to achieve mathematically optimal outcomes, decisions must be randomized. In this paper, we argue that in supervised learning settings, there exist random classifiers that perform at least as well as deterministic classifiers, and may hence be the optimal choice in many circumstances. We support our theoretical results with an empirical study indicating a positive societal attitude towards randomized artificial decision-makers, and discuss some policy and implementation issues related to the use of random classifiers that relate to and are relevant for current AI policy and standardization initiatives.