Estimating Training Data Influence by Tracing Gradient Descent

TL;DR

The paper introduces TracIn, a scalable and general method to estimate the influence of individual training examples on model predictions by tracing gradient descent, applicable across various models and tasks.

Contribution

It presents a simple, scalable influence estimation method using gradient tracing with checkpoints, applicable to any model trained with stochastic gradient descent.

Findings

TracIn accurately estimates training data influence on predictions.

The method is easy to implement with existing training infrastructure.

Applicable across diverse models and domains.

Abstract

We introduce a method called TracIn that computes the influence of a training example on a prediction made by the model. The idea is to trace how the loss on the test point changes during the training process whenever the training example of interest was utilized. We provide a scalable implementation of TracIn via: (a) a first-order gradient approximation to the exact computation, (b) saved checkpoints of standard training procedures, and (c) cherry-picking layers of a deep neural network. In contrast with previously proposed methods, TracIn is simple to implement; all it needs is the ability to work with gradients, checkpoints, and loss functions. The method is general. It applies to any machine learning model trained using stochastic gradient descent or a variant of it, agnostic of architecture, domain and task. We expect the method to be widely useful within processes that study and improve training data.