
Lowvariance Blackbox Gradient Estimates for the PlackettLuce Distribution
Learning models with discrete latent variables using stochastic gradient...
read it

Understanding Blackbox Predictions via Influence Functions
How can we explain the predictions of a blackbox model? In this paper, ...
read it

Estimating Training Data Influence by Tracking Gradient Descent
We introduce a method called TrackIn that computes the influence of a tr...
read it

Interpreting Black Box Predictions using Fisher Kernels
Research in both machine learning and psychology suggests that salient e...
read it

FastSHAP: RealTime Shapley Value Estimation
Shapley values are widely used to explain blackbox models, but they are...
read it

Human Understandable Explanation Extraction for Blackbox Classification Models Based on Matrix Factorization
In recent years, a number of artificial intelligent services have been d...
read it

Distributionally Constrained BlackBox Stochastic Gradient Estimation and Optimization
We consider stochastic gradient estimation using only blackbox function...
read it
Explaining Neural Matrix Factorization with Gradient Rollback
Explaining the predictions of neural blackbox models is an important problem, especially when such models are used in applications where user trust is crucial. Estimating the influence of training examples on a learned neural model's behavior allows us to identify training examples most responsible for a given prediction and, therefore, to faithfully explain the output of a blackbox model. The most generally applicable existing method is based on influence functions, which scale poorly for larger sample sizes and models. We propose gradient rollback, a general approach for influence estimation, applicable to neural models where each parameter update step during gradient descent touches a smaller number of parameters, even if the overall number of parameters is large. Neural matrix factorization models trained with gradient descent are part of this model class. These models are popular and have found a wide range of applications in industry. Especially knowledge graph embedding methods, which belong to this class, are used extensively. We show that gradient rollback is highly efficient at both training and test time. Moreover, we show theoretically that the difference between gradient rollback's influence approximation and the true influence on a model's behavior is smaller than known bounds on the stability of stochastic gradient descent. This establishes that gradient rollback is robustly estimating example influence. We also conduct experiments which show that gradient rollback provides faithful explanations for knowledge base completion and recommender datasets.
READ FULL TEXT
Comments
There are no comments yet.