Disentangling Influence: Using Disentangled Representations to Audit Model Predictions
By Charles Marx, Richard L. Phillips, Sorelle A. Friedler, Carlos Eduardo Scheidegger, Suresh Venkatasubramanian. 2019
26 citations recorded by the source corpus.
Motivated by the need to audit complex and black box models, there has been extensive research on quantifying how data features influence model predictions. Feature influence can be direct (a direct influence on model outcomes) and indirect (model outcomes are influenced via proxy features). Feature influence can also be expressed in aggregate over the training or test data or locally with respect to a single point. Current research has typically focused on one of each of these dimensions. In this paper, we develop disentangled influence audits, a procedure to audit the indirect influence of features. Specifically, we show that disentangled representations provide a mechanism to identify proxy features in the dataset, while allowing an explicit computation of feature influence on either individual outcomes or aggregate-level outcomes. We show through both theory and experiments that disentangled influence audits can both detect proxy features and show, for each individual or in aggregate, which of these proxy features affects the classifier being audited the most. In this respect, our method is more powerful than existing methods for ascertaining feature influence.
Subfields: Artificial Intelligence
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Flávio du Pin Calmon
- Adriana Romero-Soriano
- Charles Marx
- Deepak Kumar Sharma
- Charan Reddy
- Sina Honari
- Berk Ustun
- Samira Shabanian
Citations in the corpus
- Predictive Multiplicity in Classification (2019, methodology): “Reporting ambiguity and discrepancy also help us calibrate trust in downstream processes in the modern machine learning life-cycle (e.g., evaluating feature influence as in Kumar et al., 2020; Marx et al., 2019).”
- Benchmarking Bias Mitigation Algorithms in Representation Learning through Fairness Metrics (2021, background): “In-processing techniques [6, 7, 8, 9, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33] try to remove sensitive information from the learned representation space.”