NIPS 2018 Expo Demo

Dec. 2, 2018

Expo Schedule »

Automatic Generation of Factsheets for Trusted AI in a Runtime Environment

Sponsor: IBM Research AI

Organizers:
Rachel Belamy (IBM Research AI), Aleksandra Mojsilovic (IBM Research AI), David Piorkowski (IBM Research AI), Jason Tsay (IBM Research AI)

Presenters:
Kush R. Varshney (IBM Research AI), Dennis Wei (IBM Research AI), Amit Dhurandhar (IBM Research AI), Karthikeyan Natesan Ramamurthy (IBM Research AI), Jason Tsay (IBM Research AI)

https://www.research.ibm.com/artificial-intelligence/
Abstract:

Concerns about safety, transparency, and bias in AI are widespread, and it is easy to see how they erode trust in these systems. Part of the problem is a lack of standard practices to document how an AI service was created, tested, trained, deployed, and evaluated; how it should operate; and how it should (and should not) be used. To address this need, we recently proposed the concept of factsheets for AI services. In our paper, we argue that a Supplier’s Declaration of Conformity (SDoC, or factsheet, for short) be completed and voluntarily released by AI service developers and providers to increase the transparency of their services and engender trust in them.

Several elements or pillars form the basis for trusted AI systems.

  • Fairness: AI systems should use training data and models that are free of bias, to avoid unfair treatment of certain groups.
  • Robustness: AI systems should be safe and secure, not vulnerable to tampering or compromising the data they are trained on.
  • Explainability: AI systems should provide decisions or suggestions that can be understood by their users and developers.
  • Lineage: AI systems should include details of their development, deployment, and maintenance so they can be audited throughout their lifecycle.

We all agree that these pillars are critical. Yet, to achieve trust in AI, making progress on these issues will not be enough; it must be accompanied with the ability to measure and communicate the performance levels of a system on each of these dimensions. One way to accomplish this is to provide such information via SDoCs or factsheets for AI services.

At the NIPS expo, we will demonstrate how an SDoC or factsheet can be automatically generated in a runtime environment (such as Watson Machine Learning) and will describe the underlying algorithms.