Beyond NaN: Resiliency of Optimization Layers in the Face of Infeasibility

Authors

  • Wai Tuck Wong Singapore Management University
  • Sarah Kinsey University of Oregon
  • Ramesha Karunasena Singapore Management University
  • Thanh H. Nguyen University of Oregon
  • Arunesh Sinha Rutgers University

DOI:

https://doi.org/10.1609/aaai.v37i12.26778

Keywords:

General

Abstract

Prior work has successfully incorporated optimization layers as the last layer in neural networks for various problems, thereby allowing joint learning and planning in one neural network forward pass. In this work, we identify a weakness in such a set-up where inputs to the optimization layer lead to undefined output of the neural network. Such undefined decision outputs can lead to possible catastrophic outcomes in critical real time applications. We show that an adversary can cause such failures by forcing rank deficiency on the matrix fed to the optimization layer which results in the optimization failing to produce a solution. We provide a defense for the failure cases by controlling the condition number of the input matrix. We study the problem in the settings of synthetic data, Jigsaw Sudoku, and in speed planning for autonomous driving. We show that our proposed defense effectively prevents the framework from failing with undefined output. Finally, we surface a number of edge cases which lead to serious bugs in popular optimization solvers which can be abused as well.

Downloads

Published

2023-06-26

How to Cite

Wong, W. T., Kinsey, S., Karunasena, R., Nguyen, T. H., & Sinha, A. (2023). Beyond NaN: Resiliency of Optimization Layers in the Face of Infeasibility. Proceedings of the AAAI Conference on Artificial Intelligence, 37(12), 15242-15250. https://doi.org/10.1609/aaai.v37i12.26778

Issue

Section

AAAI Special Track on Safe and Robust AI