In real-world systems, models are frequently updated as more data becomes available, and in addition to achieving high accuracy, the goal is to also maintain a low difference in predictions compared to the base model (i.e. predictive ``churn''). If model retraining results in vastly different behavior, then it could cause negative effects in downstream systems, especially if this churn can be avoided with limited impact on model accuracy. In this paper, we show an equivalence between training with distillation using the base model as the teacher and training with an explicit constraint on the predictive churn. We then show that distillation performs strongly for low churn training against a number of recent baselines on a wide range of datasets and model architectures, including fully-connected networks, convolutional networks, and transformers.
We consider learning to optimize a classification metric defined by a black-box function of the confusion matrix. Such black-box learning settings are ubiquitous, for example, when the learner only has query access to the metric of interest, or in noisy-label and domain adaptation applications where the learner must evaluate the metric via performance evaluation using a small validation sample. Our approach is to adaptively learn example weights on the training dataset such that the resulting weighted objective best approximates the metric on the validation sample. We show how to model and estimate the example weights and use them to iteratively post-shift a pre-trained class probability estimator to construct a classifier. We also analyze the resulting procedure's statistical properties. Experiments on various label noise, domain shift, and fair classification setups confirm that our proposal is better than the individual state-of-the-art baselines for each application.
Distillation is the technique of training a "student" model based on examples that are labeled by a separate "teacher" model, which itself is trained on a labeled dataset. The most common explanations for why distillation "works" are predicated on the assumption that student is provided with \emph{soft} labels, \eg probabilities or confidences, from the teacher model. In this work, we show, that, even when the teacher model is highly overparameterized, and provides \emph{hard} labels, using a very large held-out unlabeled dataset to train the student model can result in a model that outperforms more "traditional" approaches. Our explanation for this phenomenon is based on recent work on "double descent". It has been observed that, once a model's complexity roughly exceeds the amount required to memorize the training data, increasing the complexity \emph{further} can, counterintuitively, result in \emph{better} generalization. Researchers have identified several settings in which it takes place, while others have made various attempts to explain it (thus far, with only partial success). In contrast, we avoid these questions, and instead seek to \emph{exploit} this phenomenon by demonstrating that a highly-overparameterized teacher can avoid overfitting via double descent, while a student trained on a larger independent dataset labeled by this teacher will avoid overfitting due to the size of its training set.
Metric elicitation is a recent framework for eliciting performance metrics that best reflect implicit user preferences. This framework enables a practitioner to adjust the performance metrics based on the application, context, and population at hand. However, available elicitation strategies have been limited to linear (or fractional-linear) functions of predictive rates. In this paper, we develop an approach to elicit from a wider range of complex multiclass metrics defined by quadratic functions of rates by exploiting their local linear structure. We apply this strategy to elicit quadratic metrics for group-based fairness, and also discuss how it can be generalized to higher-order polynomials. Our elicitation strategies require only relative preference feedback and are robust to both feedback and finite sample noise.
What is a fair performance metric? We consider the choice of fairness metrics through the lens of metric elicitation -- a principled framework for selecting performance metrics that best reflect implicit preferences. The use of metric elicitation enables a practitioner to tune the performance and fairness metrics to the task, context, and population at hand. Specifically, we propose a novel strategy to elicit fair performance metrics for multiclass classification problems with multiple sensitive groups that also includes selecting the trade-off between performance and fairness. The proposed elicitation strategy requires only relative preference feedback and is robust to both finite sample and feedback noise.
Many existing fairness criteria for machine learning involve equalizing or achieving some metric across \textit{protected groups} such as race or gender groups. However, practitioners trying to audit or enforce such group-based criteria can easily face the problem of noisy or biased protected group information. We study this important practical problem in two ways. First, we study the consequences of na{\"i}vely only relying on noisy protected groups: we provide an upper bound on the fairness violations on the true groups $G$ when the fairness criteria are satisfied on noisy groups $\hat{G}$. Second, we introduce two new approaches using robust optimization that, unlike the na{\"i}ve approach of only relying on $\hat{G}$, are guaranteed to satisfy fairness criteria on the true protected groups $G$ while minimizing a training objective. We provide theoretical guarantees that one such approach converges to an optimal feasible solution. Using two case studies, we empirically show that the robust approaches achieve better true group fairness guarantees than the na{\"i}ve approach.
We address the problem of training models with black-box and hard-to-optimize metrics by expressing the metric as a monotonic function of a small number of easy-to-optimize surrogates. We pose the training problem as an optimization over a relaxed surrogate space, which we solve by estimating local gradients for the metric and performing inexact convex projections. We analyze gradient estimates based on finite differences and local linear interpolations, and show convergence of our approach under smoothness assumptions with respect to the surrogates. Experimental results on classification and ranking problems verify the proposal performs on par with methods that know the mathematical formulation, and adds notable value when the form of the metric is unknown.
We present a general framework for solving a large class of learning problems with non-linear functions of classification rates. This includes problems where one wishes to optimize a non-decomposable performance metric such as the F-measure or G-mean, and constrained training problems where the classifier needs to satisfy non-linear rate constraints such as predictive parity fairness, distribution divergences or churn ratios. We extend previous two-player game approaches for constrained optimization to a game between three players to decouple the classifier rates from the non-linear objective, and seek to find an equilibrium of the game. Our approach generalizes many existing algorithms, and makes possible new algorithms with more flexibility and tighter handling of non-linear rate constraints. We provide convergence guarantees for convex functions of rates, and show how our methodology can be extended to handle sums of ratios of rates. Experiments on different fairness tasks confirm the efficacy of our approach.
We present pairwise metrics of fairness for ranking and regression models that form analogues of statistical fairness notions such as equal opportunity or equal accuracy, as well as statistical parity. Our pairwise formulation supports both discrete protected groups, and continuous protected attributes. We show that the resulting training problems can be efficiently and effectively solved using constrained optimization and robust optimization techniques based on two player game algorithms developed for fair classification. Experiments illustrate the broad applicability and trade-offs of these methods.
Designing an auction that maximizes expected revenue is an intricate task. Indeed, as of today--despite major efforts and impressive progress over the past few years--only the single-item case is fully understood. In this work, we initiate the exploration of the use of tools from deep learning on this topic. The design objective is revenue optimal, dominant-strategy incentive compatible auctions. We show that multi-layer neural networks can learn almost-optimal auctions for settings for which there are analytical solutions, such as Myerson's auction for a single item, Manelli and Vincent's mechanism for a single bidder with additive preferences over two items, or Yao's auction for two additive bidders with binary support distributions and multiple items, even if no prior knowledge about the form of optimal auctions is encoded in the network and the only feedback during training is revenue and regret. We further show how characterization results, even rather implicit ones such as Rochet's characterization through induced utilities and their gradients, can be leveraged to obtain more precise fits to the optimal design. We conclude by demonstrating the potential of deep learning for deriving optimal auctions with high revenue for poorly understood problems.