Zeno: An Interactive Framework for Behavioral Evaluation of Machine Learning

02/09/2023
by   Ángel Alexander Cabrera, et al.
0

Machine learning models with high accuracy on test data can still produce systematic failures, such as harmful biases and safety issues, when deployed in the real world. To detect and mitigate such failures, practitioners run behavioral evaluation of their models, checking model outputs for specific types of inputs. Behavioral evaluation is important but challenging, requiring that practitioners discover real-world patterns and validate systematic failures. We conducted 18 semi-structured interviews with ML practitioners to better understand the challenges of behavioral evaluation and found that it is a collaborative, use-case-first process that is not adequately supported by existing task- and domain-specific tools. Using these findings, we designed Zeno, a general-purpose framework for visualizing and testing AI systems across diverse use cases. In four case studies with participants using Zeno on real-world models, we found that practitioners were able to reproduce previous manual analyses and discover new systematic failures.

READ FULL TEXT

page 7

page 11

research
09/23/2021

Discovering and Validating AI Errors With Crowdsourced Failure Reports

AI systems can fail to learn important behaviors, leading to real-world ...
research
06/21/2023

Mass-Producing Failures of Multimodal Systems with Language Models

Deployed multimodal systems can fail in ways that evaluators did not ant...
research
01/13/2023

MLOps: A Primer for Policymakers on a New Frontier in Machine Learning

This chapter is written with the Data Scientist or MLOps professional in...
research
06/18/2021

Planning for Natural Language Failures with the AI Playbook

Prototyping AI user experiences is challenging due in part to probabilis...
research
04/12/2023

Angler: Helping Machine Translation Practitioners Prioritize Model Improvements

Machine learning (ML) models can fail in unexpected ways in the real wor...
research
07/11/2023

Empowering Cross-lingual Behavioral Testing of NLP Models with Typological Features

A challenge towards developing NLP systems for the world's languages is ...

Please sign up or login with your details

Forgot password? Click here to reset