Abstract:Learning correlations from data forms the foundation of today's machine learning (ML) and artificial intelligence (AI) research. While such an approach enables the automatic discovery of patterned relationships within big data corpora, it is susceptible to failure modes when unintended correlations are captured. This vulnerability has expanded interest in interrogating spuriousness, often critiqued as an impediment to model performance, fairness, and robustness. In this article, we trace deviations from the conventional definition of statistical spuriousness-which denotes a non-causal observation arising from either coincidence or confounding variables-to articulate how ML researchers make sense of spuriousness in practice. Drawing on a broad survey of ML literature, we conceptualize the "multiple dimensions of spuriousness," encompassing: relevance ("Models should only use correlations that are relevant to the task."), generalizability ("Models should only use correlations that generalize to unseen data"), human-likeness ("Models should only use correlations that a human would use to perform the same task"), and harmfulness ("Models should only use correlations that are not harmful"). These dimensions demonstrate that ML spuriousness goes beyond the causal/non-causal dichotomy and that the disparate interpretative paths researchers choose could meaningfully influence the trajectory of ML development. By underscoring how a fundamental problem in ML is contingently negotiated in research contexts, we contribute to ongoing debates about responsible practices in AI development.
Abstract:Neural networks can fail when the data contains spurious correlations. To understand this phenomenon, researchers have proposed numerous spurious correlations benchmarks upon which to evaluate mitigation methods. However, we observe that these benchmarks exhibit substantial disagreement, with the best methods on one benchmark performing poorly on another. We explore this disagreement, and examine benchmark validity by defining three desiderata that a benchmark should satisfy in order to meaningfully evaluate methods. Our results have implications for both benchmarks and mitigations: we find that certain benchmarks are not meaningful measures of method performance, and that several methods are not sufficiently robust for widespread use. We present a simple recipe for practitioners to choose methods using the most similar benchmark to their given problem.
Abstract:Rapid progress in text-to-image generative models coupled with their deployment for visual content creation has magnified the importance of thoroughly evaluating their performance and identifying potential biases. In pursuit of models that generate images that are realistic, diverse, visually appealing, and consistent with the given prompt, researchers and practitioners often turn to automated metrics to facilitate scalable and cost-effective performance profiling. However, commonly-used metrics often fail to account for the full diversity of human preference; often even in-depth human evaluations face challenges with subjectivity, especially as interpretations of evaluation criteria vary across regions and cultures. In this work, we conduct a large, cross-cultural study to study how much annotators in Africa, Europe, and Southeast Asia vary in their perception of geographic representation, visual appeal, and consistency in real and generated images from state-of-the art public APIs. We collect over 65,000 image annotations and 20 survey responses. We contrast human annotations with common automated metrics, finding that human preferences vary notably across geographic location and that current metrics do not fully account for this diversity. For example, annotators in different locations often disagree on whether exaggerated, stereotypical depictions of a region are considered geographically representative. In addition, the utility of automatic evaluations is dependent on assumptions about their set-up, such as the alignment of feature extractors with human perception of object similarity or the definition of "appeal" captured in reference datasets used to ground evaluations. We recommend steps for improved automatic and human evaluations.
Abstract:The simple idea that not all things are equally difficult has surprising implications when applied in a fairness context. In this work we explore how "difficulty" is model-specific, such that different models find different parts of a dataset challenging. When difficulty correlates with group information, we term this difficulty disparity. Drawing a connection with recent work exploring the inductive bias towards simplicity of SGD-trained models, we show that when such a disparity exists, it is further amplified by commonly-used models. We quantify this amplification factor across a range of settings aiming towards a fuller understanding of the role of model bias. We also present a challenge to the simplifying assumption that "fixing" a dataset is sufficient to ensure unbiased performance.
Abstract:Amid mounting concern about the reliability and credibility of machine learning research, we present a principled framework for making robust and generalizable claims: the Multiverse Analysis. Our framework builds upon the Multiverse Analysis (Steegen et al., 2016) introduced in response to psychology's own reproducibility crisis. To efficiently explore high-dimensional and often continuous ML search spaces, we model the multiverse with a Gaussian Process surrogate and apply Bayesian experimental design. Our framework is designed to facilitate drawing robust scientific conclusions about model performance, and thus our approach focuses on exploration rather than conventional optimization. In the first of two case studies, we investigate disputed claims about the relative merit of adaptive optimizers. Second, we synthesize conflicting research on the effect of learning rate on the large batch training generalization gap. For the machine learning community, the Multiverse Analysis is a simple and effective technique for identifying robust claims, for increasing transparency, and a step toward improved reproducibility.
Abstract:We investigate the effect of task ordering on continual learning performance. We conduct an extensive series of empirical experiments on synthetic and naturalistic datasets and show that reordering tasks significantly affects the amount of catastrophic forgetting. Connecting to the field of curriculum learning, we show that the effect of task ordering can be exploited to modify continual learning performance, and present a simple approach for doing so. Our method computes the distance between all pairs of tasks, where distance is defined as the source task curvature of a gradient step toward the target task. Using statistically rigorous methods and sound experimental design, we show that task ordering is an important aspect of continual learning that can be modified for improved performance.
Abstract:In this paper we explore whether the fundamental tool of experimental psychology, the behavioral experiment, has the power to generate insight not only into humans and animals, but artificial systems too. We apply the techniques of experimental psychology to investigating catastrophic forgetting in neural networks. We present a series of controlled experiments with two-layer ReLU networks, and exploratory results revealing a new understanding of the behavior of catastrophic forgetting. Alongside our empirical findings, we demonstrate an alternative, behavior-first approach to investigating neural network phenomena.
Abstract:In the early 2010s, a crisis of reproducibility rocked the field of psychology. Following a period of reflection, the field has responded with radical reform of its scientific practices. More recently, similar questions about the reproducibility of machine learning research have also come to the fore. In this short paper, we present select ideas from psychology's reformation, translating them into relevance for a machine learning audience.