In Reinforcement Learning from Human Feedback (RLHF), it is crucial to learn suitable reward models from human feedback to align large language models (LLMs) with human intentions.However, human feedback can often be noisy, inconsistent, or biased, especially when evaluating complex responses.Such feedback can lead to misaligned reward signals, potentially causing unintended side effects during the RLHF process.To address these challenges, we explore the use of influence functions to measure the impact of human feedback on the performance of reward models.We propose a compute-efficient approximation method that enables the application of influence functions to LLM-based reward models and large-scale preference datasets.Our experiments showcase two key applications of influence functions: (1) detecting common labeler biases in human feedback datasets and (2) guiding labelers in refining their strategies to better align with expert feedback.By quantifying the impact of human feedback, we believe that influence functions can enhance feedback interpretability and contribute to scalable oversight in RLHF, helping labelers provide more accurate and consistent feedback.Source code is available at https://github.com/mintaywon/IF_RLHF.

Evaluating the contribution of individual data points to a model's prediction is critical for interpreting model predictions and improving model performance. Existing data contribution methods have been applied to various data types, including tabular data, images, and text; however, their primary focus has been on i.i.d. settings. Despite the pressing need for principled approaches tailored to time series datasets, the problem of estimating data contribution in such settings remains under-explored, possibly due to challenges associated with handling inherent temporal dependencies. This paper introduces TimeInf, a model-agnostic data contribution estimation method for time-series datasets. By leveraging influence scores, TimeInf attributes model predictions to individual time points while preserving temporal structures between the time points. Our empirical results show that TimeInf effectively detects time series anomalies and outperforms existing data attribution techniques as well as state-of-the-art anomaly detection methods. Moreover, TimeInf offers interpretable attributions of data values, allowing us to distinguish diverse anomalous patterns through visualizations. We also showcase a potential application of TimeInf in identifying mislabeled anomalies in the ground truth annotations.

Machine unlearning focuses on the computationally efficient removal of specific training data from trained models, ensuring that the influence of forgotten data is effectively eliminated without the need for full retraining. Despite advances in low-dimensional settings, where the number of parameters \( p \) is much smaller than the sample size \( n \), extending similar theoretical guarantees to high-dimensional regimes remains challenging. We study an unlearning algorithm that starts from the original model parameters and performs a theory-guided sequence of Newton steps. After this update, carefully scaled isotropic Laplacian noise is added to the estimate to ensure that any (potential) residual influence of the deletion set is completely removed. We show that when both \( n, p \to \infty \) with a fixed ratio \( n/p \), significant theoretical and computational obstacles arise due to the interplay between the complexity of the model and the finite signal-to-noise ratio. Finally, we show that, unlike in low-dimensional settings where one Newton step suffices, in high-dimensional problems at least two Newton steps are required to effectively unlearn a fixed number of data points, and even more steps are required when the deletion set scales with $n$. We provide numerical experiments to support the theoretical claims of the paper.

Quantifying the impact of training data points is crucial for understanding the outputs of machine learning models and for improving the transparency of the AI pipeline. The influence function is a principled and popular data attribution method, but its computational cost often makes it challenging to use. This issue becomes more pronounced in the setting of large language models and text-to-image models. In this work, we propose DataInf, an efficient influence approximation method that is practical for large-scale generative AI models. Leveraging an easy-to-compute closed-form expression, DataInf outperforms existing influence computation algorithms in terms of computational and memory efficiency. Our theoretical analysis shows that DataInf is particularly well-suited for parameter-efficient fine-tuning techniques such as LoRA. Through systematic empirical evaluations, we show that DataInf accurately approximates influence scores and is orders of magnitude faster than existing methods. In applications to RoBERTa-large, Llama-2-13B-chat, and stable-diffusion-v1.5 models, DataInf effectively identifies the most influential fine-tuning examples better than other approximate influence scores. Moreover, it can help to identify which data points are mislabeled.

Data Shapley provides a principled approach to data valuation and plays a crucial role in data-centric machine learning (ML) research. Data selection is considered a standard application of Data Shapley. However, its data selection performance has shown to be inconsistent across settings in the literature. This study aims to deepen our understanding of this phenomenon. We introduce a hypothesis testing framework and show that Data Shapley’s performance can be no better than random selection without specific constraints on utility functions. We identify a class of utility functions, monotonically transformed modular functions, within which Data Shapley optimally selects data. Based on this insight, we propose a heuristic for predicting Data Shapley’s effectiveness in data selection tasks. Our experiments corroborate these findings, adding new insights into when Data Shapley may or may not succeed.

Data valuation has emerged as a powerful framework for quantifying each datum's contribution to the training of a machine learning model. However, it is crucial to recognize that the quality of cells within a single data point can vary greatly in practice. For example, even in the case of an abnormal data point, not all cells are necessarily noisy. The single scalar score assigned by existing data valuation methods blurs the distinction between noisy and clean cells of a data point, making it challenging to interpret the data values. In this paper, we propose 2D-OOB, an out-of-bag estimation framework for jointly determining helpful (or detrimental) samples as well as the particular cells that drive them. Our comprehensive experiments demonstrate that 2D-OOB achieves state-of-the-art performance across multiple use cases while being exponentially faster. Specifically, 2D-OOB shows promising results in detecting and rectifying fine-grained outliers at the cell level, and localizing backdoor triggers in data poisoning attacks.

Data valuation is a powerful framework for providing statistical insights into which data are beneficial or detrimental to model training. Many Shapley-based data valuation methods have shown promising results in various downstream tasks, however, they are well known to be computationally challenging as it requires training a large number of models. As a result, it has been recognized as infeasible to apply to large datasets. To address this issue, we propose Data-OOB, a new data valuation method for a bagging model that utilizes the out-of-bag estimate. The proposed method is computationally efficient and can scale to millions of data by reusing trained weak learners. Specifically, Data-OOB takes less than $2.25$ hours on a single CPU processor when there are $10^6$ samples to evaluate and the input dimension is $100$. Furthermore, Data-OOB has solid theoretical interpretations in that it identifies the same important data point as the infinitesimal jackknife influence function when two different points are compared. We conduct comprehensive experiments using 12 classification datasets, each with thousands of sample sizes. We demonstrate that the proposed method significantly outperforms existing state-of-the-art data valuation methods in identifying mislabeled data and finding a set of helpful (or harmful) data points, highlighting the potential for applying data values in real-world applications.

Understanding the performance of machine learning (ML) models across diverse data distributions is critically important for reliable applications. Despite recent empirical studies positing a near-perfect linear correlation between in-distribution (ID) and out-of-distribution (OOD) accuracies, we empirically demonstrate that this correlation is more nuanced under subpopulation shifts. Through rigorous experimentation and analysis across a variety of datasets, models, and training epochs, we demonstrate that OOD performance often has a nonlinear correlation with ID performance in subpopulation shifts. Our findings, which contrast previous studies that have posited a linear correlation in model performance during distribution shifts, reveal a "moon shape" correlation (parabolic uptrend curve) between the test performance on the majority subpopulation and the minority subpopulation. This non-trivial nonlinear correlation holds across model architectures, hyperparameters, training durations, and the imbalance between subpopulations. Furthermore, we found that the nonlinearity of this "moon shape" is causally influenced by the degree of spurious correlations in the training data. Our controlled experiments show that stronger spurious correlation in the training data creates more nonlinear performance correlation. We provide complementary experimental and theoretical analyses for this phenomenon, and discuss its implications for ML reliability and fairness. Our work highlights the importance of understanding the nonlinear effects of model improvement on performance in different subpopulations, and has the potential to inform the development of more equitable and responsible machine learning models.

Assessing the quality and impact of individual data points is critical for improving model performance and mitigating undesirable biases within the training dataset. Several data valuation algorithms have been proposed to quantify data quality, however, there lacks a systemic and standardized benchmarking system for data valuation. In this paper, we introduce OpenDataVal, an easy-to-use and unified benchmark framework that empowers researchers and practitioners to apply and compare various data valuation algorithms. OpenDataVal provides an integrated environment that includes (i) a diverse collection of image, natural language, and tabular datasets, (ii) implementations of eleven different state-of-the-art data valuation algorithms, and (iii) a prediction model API that can import any models in scikit-learn. Furthermore, we propose four downstream machine learning tasks for evaluating the quality of data values. We perform benchmarking analysis using OpenDataVal, quantifying and comparing the efficacy of state-of-the-art data valuation approaches. We find that no single algorithm performs uniformly best across all tasks, and an appropriate algorithm should be employed for a user's downstream task. OpenDataVal is publicly available at https://opendataval.github.io with comprehensive documentation. Furthermore, we provide a leaderboard where researchers can evaluate the effectiveness of their own data valuation algorithms.

Data Shapley has recently been proposed as a principled framework to quantify the contribution of individual datum in machine learning. It can effectively identify helpful or harmful data points for a learning algorithm. In this paper, we propose Beta Shapley, which is a substantial generalization of Data Shapley. Beta Shapley arises naturally by relaxing the efficiency axiom of the Shapley value, which is not critical for machine learning settings. Beta Shapley unifies several popular data valuation methods and includes data Shapley as a special case. Moreover, we prove that Beta Shapley has several desirable statistical properties and propose efficient algorithms to estimate it. We demonstrate that Beta Shapley outperforms state-of-the-art data valuation methods on several downstream ML tasks such as: 1) detecting mislabeled training data; 2) learning with subsamples; and 3) identifying points whose addition or removal have the largest positive or negative impact on the model.

Shapley value is a popular approach for measuring the influence of individual features. While Shapley feature attribution is built upon desiderata from game theory, some of its constraints may be less natural in certain machine learning settings, leading to unintuitive model interpretation. In particular, the Shapley value uses the same weight for all marginal contributions---i.e. it gives the same importance when a large number of other features are given versus when a small number of other features are given. This property can be problematic if larger feature sets are more or less informative than smaller feature sets. Our work performs a rigorous analysis of the potential limitations of Shapley feature attribution. We identify simple settings where the Shapley value is mathematically suboptimal by assigning larger attributions for less influential features. Motivated by this observation, we propose WeightedSHAP, which generalizes the Shapley value and learns which marginal contributions to focus directly from data. On several real-world datasets, we demonstrate that the influential features identified by WeightedSHAP are better able to recapitulate the model's predictions compared to the features identified by the Shapley value.

We present modality gap, an intriguing geometric phenomenon of the representation space of multi-modal models. Specifically, we show that different data modalities (e.g. images and text) are embedded at arm's length in their shared representation in multi-modal models such as CLIP. Our systematic analysis demonstrates that this gap is caused by a combination of model initialization and contrastive learning optimization. In model initialization, we show empirically and theoretically that the representation of a common deep neural network is restricted to a narrow cone. As a consequence, in a multi-modal model with two encoders, the representations of the two modalities are clearly apart when the model is initialized. During optimization, contrastive learning keeps the different modalities separate by a certain distance, which is influenced by the temperature parameter in the loss function. Our experiments further demonstrate that varying the modality gap distance has a significant impact in improving the model's downstream zero-shot classification performance and fairness.

This papers studies how competition affects machine learning (ML) predictors. As ML becomes more ubiquitous, it is often deployed by companies to compete over customers. For example, digital platforms like Yelp use ML to predict user preference and make recommendations. A service that is more often queried by users, perhaps because it more accurately anticipates user preferences, is also more likely to obtain additional user data (e.g. in the form of a Yelp review). Thus, competing predictors cause feedback loops whereby a predictor’s performance impacts what training data it receives and biases its predictions over time. We introduce a flexible model of competing ML predictors that enables both rapid experimentation and theoretical tractability. We show with empirical and mathematical analysis that competition causes predictors to specialize for specific sub-populations at the cost of worse performance over the general population. We further analyze the impact of predictor specialization on the overall prediction quality experienced by users. We show that having too few or too many competing predictors in a market can hurt the overall prediction quality. Our theory is complemented by experiments on several real datasets using popular learning algorithms, such as neural networks and nearest neighbor methods.

Distributional data Shapley value (DShapley) has recently been proposed as a principled framework to quantify the contribution of individual datum in machine learning. DShapley develops the foundational game theory concept of Shapley values into a statistical framework and can be applied to identify data points that are useful (or harmful) to a learning algorithm. Estimating DShapley is computationally expensive, however, and this can be a major challenge to using it in practice. Moreover, there has been little mathematical analyses of how this value depends on data characteristics. In this paper, we derive the first analytic expressions for DShapley for the canonical problems of linear regression, binary classification, and non-parametric density estimation. These analytic forms provide new algorithms to estimate DShapley that are several orders of magnitude faster than previous state-of-the-art methods. Furthermore, our formulas are directly interpretable and provide quantitative insights into how the value varies for different types of data. We demonstrate the practical efficacy of our approach on multiple real and synthetic datasets.

Anomaly detection is the task of finding abnormal data that are distinct from normal behavior. Current deep learning-based anomaly detection methods train neural networks with normal data alone and calculate anomaly scores based on the trained model. In this work, we formalize current practices, build a theoretical framework of anomaly detection algorithms equipped with an objective function and a hypothesis space, and establish a desirable property of the anomaly detection algorithm, namely, admissibility. Admissibility implies that optimal autoencoders for normal data yield a larger reconstruction error for anomalous data than that for normal data on average. We then propose a class of admissible anomaly detection algorithms equipped with an integral probability metric-based objective function and a class of autoencoders, Lipschitz continuous autoencoders. The proposed algorithm for Wasserstein distance is implemented by minimizing an approximated Wasserstein distance with a penalty to enforce Lipschitz continuity with respect to Wasserstein distance. Through ablation studies, we demonstrate the efficacy of enforcing Lipschitz continuity of the proposed method. The proposed method is shown to be more effective in detecting anomalies than existing methods via applications to network traffic and image datasets.

Wasserstein distributionally robust optimization (WDRO) attempts to learn a model that minimizes the local worst-case risk in the vicinity of the empirical data distribution defined by Wasserstein ball. While WDRO has received attention as a promising tool for inference since its introduction, its theoretical understanding has not been fully matured. Gao et al. (2017) proposed a minimizer based on a tractable approximation of the local worst-case risk, but without showing risk consistency. In this paper, we propose a minimizer based on a novel approximation theorem and provide the corresponding risk consistency results. Furthermore, we develop WDRO inference for locally perturbed data that include the Mixup (Zhang et al., 2017) as a special case. We show that our approximation and risk consistency results naturally extend to the cases when data are locally perturbed. Numerical experiments demonstrate robustness of the proposed method using image classification datasets. Our results show that the proposed method achieves significantly higher accuracy than baseline models on noisy datasets.