~/talk

A developer's guide to responsible AI review processes

on YouTube 19m
Talkat Google I/O 2022 · 2022-05-11..05-12· [machine-learning, engineering-leadership]

speakers

name:Madeleine Elish
links:none on file
name:Jen Gennai
links:none on file

resources

video:https://youtu.be/wvytEFQ7mIo
slides:none on file

summary

Presents a concrete three-step process—Analyze, Adjust, Integrate—and a set of open source tools that teams actually use: TensorFlow Responsible AI Toolkit, Know Your Data (interactive exploration of ~70 datasets), Fairness Indicators, LIT, MinDiff and a new Counterfactuals remediation, plus Model Cards and a TFX component to auto-generate them. Describes COVID‑19 Public Forecasts (launched, monitored, then wound down when data quality fell) and a celebrity recognition API improved after three fairness tests across gender, age and skin tone.

AI-written summary · about

chapters (7)

  1. 0:00Intro
  2. 0:16Overview
  3. 4:10Analyze
  4. 5:53Resources
  5. 7:27Adjust
  6. 9:53Integrate
  7. 14:00Conclusion

related talks

~/search
type to search 6,249 conferences, 138,987 speakers and 185,172 talks
~/help
cmdctrl k
search every conference, speaker and series
/
focus the filter search
h
home
c
conferences
s
speakers
e
series
t
light / dark — the status bar button does the same
?
this help
esc
close this, then clear every filter — or just the search, while you are typing in it

Ignored while you are typing. Pages without a filter rail have no search to focus.