loading page

ProactiV: Studying deep learning model behavior under input transformations
  • +1
  • Vidya Prasad ,
  • Ruud van Sloun ,
  • Anna Vilanova ,
  • Nicola Pezzotti
Vidya Prasad
Author Profile
Ruud van Sloun
Author Profile
Anna Vilanova
Author Profile
Nicola Pezzotti
Author Profile

Abstract

Deep learning (DL) models have shown performance benefits across many applications, from classification to, more recently, image-to-image translation. However, low interpretability often leads to unexpected model behavior when deployed in the real world. Usually, this unexpected behavior is because the training data domain does not reflect the deployment data domain. Identifying model breaking points under input conditions and domain shifts, i.e., input transformations, is essential to improve models. Although visual analytics (VA) has shown promise in studying the behavior of model outputs under continually varying inputs, existing methods mainly focus on per class or instance-level analysis. We aim to generalize beyond classification where classes do not exist and provide a global holistic model behavior overview under co-occurring input transformations. We present a DL model-agnostic VA method (ProactiV ) to help model developers proactively study output behavior under input transformations to identify and verify breakpoints. ProactiV relies on a proposed input optimization method to determine how far a given transformed input has to move in the input data manifold to achieve the desired output. The data from this optimization process allows the study of global and local model output behavior under input transformations at scale. Additionally, the optimization method provides insight into the input characteristics that result in desired outputs and helps recognize model biases. We highlight how ProactiV effectively supports studying model behavior under input transformations to identify and verify model breakpoints with example classification and image-to-image translation tasks.
2023Published in IEEE Transactions on Visualization and Computer Graphics on pages 1-14. 10.1109/TVCG.2023.3301722