Influence functions (IF) have been seen as a technique for explaining model predictions through the lens of the training data.

Feature attributions attempt to highlight what inputs drive predictive power. Good attributions or explanations are thus those that produce inputs that retain this predictive power; accordingly, evaluations of explanations score their quality of prediction. However, evaluations produce scores better than what appears possible from the values in the explanation for a class of explanations, called encoding explanations. Probing for encoding remains a challenge because there is no general characterization of what gives the extra predictive power. We develop a definition of encoding that identifies this extra predictive power via conditional dependence and show that the definition fits existing examples of encoding. This definition implies, in contrast to encoding explanations, that non-encoding explanations contain all the informative inputs used to produce the explanation, giving them a "what you see is what you get" property, which makes them transparent and simple to use. Next, we prove that existing scores (ROAR, FRESH, EVAL-X) do not rank non-encoding explanations above encoding ones, and develop STRIPE-X which ranks them correctly. After empirically demonstrating the theoretical insights, we use STRIPE-X to show that despite prompting an LLM to produce non-encoding explanations for a sentiment analysis task, the LLM-generated explanations encode.

Ensembling neural networks is an effective way to increase accuracy, and can often match the performance of individual larger models. This observation poses a natural question: given the choice between a deep ensemble and a single neural network with similar accuracy, is one preferable over the other? Recent work suggests that deep ensembles may offer distinct benefits beyond predictive power: namely, uncertainty quantification and robustness to dataset shift. In this work, we demonstrate limitations to these purported benefits, and show that a single (but larger) neural network can replicate these qualities. First, we show that ensemble diversity, by any metric, does not meaningfully contribute to an ensemble's ability to detect out-of-distribution (OOD) data, but is instead highly correlated with the relative improvement of a single larger model. Second, we show that the OOD performance afforded by ensembles is strongly determined by their in-distribution (InD) performance, and - in this sense - is not indicative of any effective robustness. While deep ensembles are a practical way to achieve improvements to predictive power, uncertainty quantification, and robustness, our results show that these improvements can be replicated by a (larger) single model.

Despite their renowned in-domain predictive power, convolutional neural networks are known to rely more on high-frequency patterns that humans deem superficial than on low-frequency patterns that agree better with intuitions about what constitutes category membership. This paper proposes a method for training robust convolutional networks by penalizing the predictive power of the local representations learned by earlier layers. Intuitively, our networks are forced to discard predictive signals such as color and texture that can be gleaned from local receptive fields and to rely instead on the global structures of the image. Across a battery of synthetic and benchmark domain adaptation tasks, our method confers improved generalization out of the domain. Additionally, to evaluate cross-domain transfer, we introduce ImageNet-Sketch, a new dataset consisting of sketch-like images that matches the ImageNet classification validation set in scale and dimension.

Understanding the inner workings of complex machine learning models is a long-standing problem and most recent research has focused on local interpretability. To assess the role of individual input features in a global sense, we explore the perspective of defining feature importance through the predictive power associated with each feature. We introduce two notions of predictive power (model-based and universal) and formalize this approach with a framework of additive importance measures, which unifies numerous methods in the literature. We then propose SAGE, a model-agnostic method that quantifies predictive power while accounting for feature interactions. Our experiments show that SAGE can be calculated efficiently and that it assigns more accurate importance values than other methods.

Noisy labels are inevitable in large real-world datasets. In this work, we explore an area understudied by previous works --- how the network's architecture impacts its robustness to noisy labels. We provide a formal framework connecting the robustness of a network to the alignments between its architecture and target/noise functions. Our framework measures a network's robustness via the predictive power in its representations --- the test performance of a linear model trained on the learned representations using a small set of clean labels. We hypothesize that a network is more robust to noisy labels if its architecture is more aligned with the target function than the noise. To support our hypothesis, we provide both theoretical and empirical evidence across various neural network architectures and different domains. We also find that when the network is well-aligned with the target function, its predictive power in representations could improve upon state-of-the-art (SOTA) noisy-label-training methods in terms of test accuracy and even outperform sophisticated methods that use clean labels.

The performance of machine learning models on tabular data is critically dependent on high-quality feature engineering. While Large Language Models (LLMs) have shown promise in automating feature extraction (AutoFE), existing methods are often limited by monolithic LLM architectures, simplistic quantitative feedback, and a failure to systematically integrate external domain knowledge. This paper introduces Rogue One, a novel, LLM-based multi-agent framework for knowledge-informed automatic feature extraction. Rogue One operationalizes a decentralized system of three specialized agents-Scientist, Extractor, and Tester-that collaborate iteratively to discover, generate, and validate predictive features. Crucially, the framework moves beyond primitive accuracy scores by introducing a rich, qualitative feedback mechanism and a "flooding-pruning" strategy, allowing it to dynamically balance feature exploration and exploitation. By actively incorporating external knowledge via an integrated retrieval-augmented (RAG) system, Rogue One generates features that are not only statistically powerful but also semantically meaningful and interpretable. We demonstrate that Rogue One significantly outperforms state-of-the-art methods on a comprehensive suite of 19 classification and 9 regression datasets. Furthermore, we show qualitatively that the system surfaces novel, testable hypotheses, such as identifying a new potential biomarker in the myocardial dataset, underscoring its utility as a tool for scientific discovery.

Several recent papers considered a related question of a function's sensitivity to its various inputs

Understanding the inner workings of complex machine learning models is a longstanding problem and most recent research has focused on local interpretability.