ModelPred: A Framework for Predicting Trained Model from Training Data

Abstract

In this work, we propose ModelPred, a framework that helps to understand the impact of changes in training data on a trained model. This is critical for building trust in various stages of a machine learning pipeline: from cleaning poor-quality samples and tracking important ones to be collected during data preparation, to calibrating uncertainty of model prediction, to interpreting why certain behaviors of a model emerge during deployment. Specifically, ModelPred learns a parameterized function that takes a dataset $S$ as the input and predicts the model obtained by training on $S$. Our work differs from the recent work of Datamodels [1] as we aim for predicting the trained model parameters directly instead of the trained model behaviors. We demonstrate that a neural network-based set function class is capable of learning the complex relationships between the training data and model parameters. We introduce novel global and local regularization techniques to prevent overfitting and we rigorously characterize the expressive power of neural networks (NN) in approximating the end-to-end training process. Through extensive empirical investigations, we show that ModelPred enables a variety of applications that boost the interpretability and accountability of machine learning (ML), such as data valuation, data selection, memorization quantification, and model calibration.

References in corpus (21)

• On Calibration of Modern Neural Networks
• Understanding Black-box Predictions via Influence Functions
• Searching for Exotic Particles in High-Energy Physics with Deep Learning
• Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles
• Towards Efficient Data Valuation Based on the Shapley Value
• Learning to Learn without Gradient Descent by Gradient Descent
• Learning to Optimize
• Data Shapley: Equitable Valuation of Data for Machine Learning
• Learned Optimizers that Scale and Generalize
• On the Accuracy of Influence Functions for Measuring Group Effects
• Making AI Forget You: Data Deletion in Machine Learning
• Learning Gradient Descent: Better Generalization and Longer Horizons
• Understanding and correcting pathologies in the training of learned optimizers
• DeltaGrad: Rapid retraining of machine learning models
• A Swiss Army Infinitesimal Jackknife
• Learning to Stop While Learning to Predict
• Influence Functions in Deep Learning Are Fragile
• A Higher-Order Swiss Army Infinitesimal Jackknife
• Certifiable Machine Unlearning for Linear Models
• Machine Unlearning for Random Forests
• A Generalizable Approach to Learning Optimizers