SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2408.13648 · 2024

Explanatory Model Monitoring to Understand the Effects of Feature Shifts on Performance

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 3 functions out of this paper's own repositories and ran 2 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
thomdeck/xpe canonical 2 of 3
FunctionStatusWhere it lives
line_from_points Ran thomdeck/xpe/src/utils.py
code served (permissive licence) · get_code("c329011944ef77b9")
zigzag Ran thomdeck/xpe/src/utils.py
code served (permissive licence) · get_code("c03e099203209b39")
spatter Not yet run thomdeck/xpe/src/utils.py
code served (permissive licence) · get_code("53773ea2ad271a59")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Monitoring and maintaining machine learning models are among the most critical challenges in translating recent advances in the field into real-world applications. However, current monitoring methods lack the capability of provide actionable insights answering the question of why the performance of a particular model really degraded. In this work, we propose a novel approach to explain the behavior of a black-box model under feature shifts by attributing an estimated performance change to interpretable input characteristics. We refer to our method that combines concepts from Optimal Transport and Shapley Values as Explanatory Performance Estimation (XPE). We analyze the underlying assumptions and demonstrate the superiority of our approach over several baselines on different data sets across various data modalities such as images, audio, and tabular data. We also indicate how the generated results can lead to valuable insights, enabling explanatory model monitoring by revealing potential root causes for model deterioration and guiding toward actionable countermeasures.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2408.13648")
get_code_for_paper("2408.13648")
have("2408.13648")

Connect an agent — have() is free.