Proxima: A Proxy Model-Based Approach to Influence Analysis

Published

September 25, 2024

Author(s)

Sunny Shree, Yu Lei, Raghu Kacker, David Kuhn

Abstract

Machine learning (ML)-based Artificial Intelligence (AI) systems rely on training data to perform optimally, but the internal workings of how ML models learn from and use this data are often a black- box. Influence analysis provides valuable insights into the model's behavior by evaluating the effect of individual training instances on the model's predictions. However, calculating the influence of each training data can be computationally expensive. In this paper, we propose a proxy model-based approach to influence analysis called Proxima. The main idea of our approach is to use a subset of training instances to create a proxy model that is simpler than the original model, and then use the proxy model and the subset to perform influence analysis. We evaluate Proxima on eight ML mod- els trained using real-world datasets. We compare Proxima to two state-of-the-art influence analysis tools, i.e., FastIF and Scaling-Up. Our experimental results suggest that the proposed approach can successfully perform and speed up the influence analysis process, and, in most cases, perform better than FastIF and Scaling-Up.

Proceedings Title

IEEE International Conference on Artificial Intelligence Testing (AITest)

Conference Dates

July 15-18, 2024

Conference Location

Shanghai, CN

Conference Title

6th IEEE International Conference on Artificial Intelligence Testing

Pub Type

Conferences

Download Paper

https://doi.org/10.1109/AITest62860.2024.00016

Local Download

Keywords

Datasets, Machine Learning, Neural Networks, Influence Analysis, Influence Functions, Training Data Similarity

Citation

Shree, S. , Lei, Y. , Kacker, R. and Kuhn, D. (2024), Proxima: A Proxy Model-Based Approach to Influence Analysis, IEEE International Conference on Artificial Intelligence Testing (AITest), Shanghai, CN, [online], https://doi.org/10.1109/AITest62860.2024.00016, https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=936306 (Accessed December 12, 2024)