Search for: All records

Abstract—We present the Seldonian Toolkit, which enables software engineers to integrate provably safe and fair machine learning algorithms into their systems. Software systems that use data and machine learning are routinely deployed in a wide range of settings from medical applications, autonomous vehicles, the criminal justice system, and hiring processes. These systems, however, can produce unsafe and unfair behavior, such as suggesting potentially fatal medical treatments, making racist or sexist predictions, or facilitating radicalization and polarization. To reduce these undesirable behaviors, software engineers need the ability to easily integrate their machine- learning-based systems with domain-specific safety and fairness requirements defined by domain experts, such as doctors and hiring managers. The Seldonian Toolkit provides special machine learning algorithms that enable software engineers to incorporate such expert-defined requirements of safety and fairness into their systems, while provably guaranteeing those requirements will be satisfied. A video demonstrating the Seldonian Toolkit is available at https://youtu.be/wHR-hDm9jX4/. 

There are reinforcement learning scenarios - e.g., in medicine - where we are compelled to be as confident as possible that a policy change will result in an improvement before implementing it. In such scenarios, we can employ off-policy evaluation (OPE). The basic idea of OPE is to record histories of behaviors under the current policy, and then develop an estimate of the quality of a proposed new policy, seeing what the behavior would have been under the new policy. As we are evaluating the policy without actually using it, we have the "off-policy" of OPE. Applying a concentration inequality to the estimate, we derive a confidence interval for the expected quality of the new policy. If the confidence interval lies above that of the current policy, we can change policies with high confidence that we will do no harm. We focus here on the mathematics of this method, by mechanizing the soundness of off-policy evaluation. A natural side effect of the mechanization is both to clarify all the result’s mathematical assumptions and preconditions, and to further develop HOL4’s library of verified statistical mathematics, including concentration inequalities. Of more significance, the OPE method relies on importance sampling, whose soundness we prove using a measure-theoretic approach. In fact, we generalize the standard result, showing it for contexts comprising both discrete and continuous probability distributions. 

Recent studies found that using machine learning for social applications can lead to injustice in the form of racist, sexist, and otherwise unfair and discriminatory outcomes. To address this challenge, recent machine learning algorithms have been designed to limit the likelihood such unfair behavior occurs. However, these approaches typically assume the data used for training is representative of what will be encountered in deployment, which is often untrue. In particular, if certain subgroups of the population become more or less probable in deployment (a phenomenon we call demographic shift), prior work's fairness assurances are often invalid. In this paper, we consider the impact of demographic shift and present a class of algorithms, called Shifty algorithms, that provide high-confidence behavioral guarantees that hold under demographic shift when data from the deployment environment is unavailable during training. Shifty, the first technique of its kind, demonstrates an effective strategy for designing algorithms to overcome demographic shift's challenges. We evaluate Shifty using the UCI Adult Census dataset, as well as a real-world dataset of university entrance exams and subsequent student success. We show that the learned models avoid bias under demographic shift, unlike existing methods. Our experiments demonstrate that our algorithm's high-confidence fairness guarantees are valid in practice and that our algorithm is an effective tool for training models that are fair when demographic shift occurs. 

When faced with sequential decision-making problems, it is often useful to be able to predict what would happen if decisions were made using a new policy. Those predictions must often be based on data collected under some previously used decision-making rule. Many previous methods enable such off-policy (or counterfactual) estimation of the expected value of a performance measure called the return. In this paper, we take the first steps towards a universal off-policy estimator (UnO)—one that provides off-policy estimates and high-confidence bounds for any parameter of the return distribution. We use UnO for estimating and simultaneously bounding the mean, variance, quantiles/median, inter-quantile range, CVaR, and the entire cumulative distribution of returns. Finally, we also discuss UnO’s applicability in various settings, including fully observable, partially observable (i.e., with unobserved confounders), Markovian, non-Markovian, stationary, smoothly non-stationary, and discrete distribution shifts.