Four short links: 27 May 2020
Facebook Ethics, Ubiquitous Voice, ML in Production, and Prediction Limitations
- Facebook Reportedly Ignored Its Own Research Showing Algorithms Divided Users — “Our algorithms exploit the human brain’s attraction to divisiveness,” one slide from the presentation read. The group found that if this core element of its recommendation engine were left unchecked, it would continue to serve Facebook users “more and more divisive content in an effort to gain user attention & increase time on the platform.” A separate internal report, crafted in 2016, said 64 percent of people who joined an extremist group on Facebook only did so because the company’s algorithm recommended it to them, the WSJ reports.
- Voice in Everything — Look, my point is that this is not beyond the reach of very clever people with computers. Stick a timer in my stove, a switch in my light bulb, give each a super limited vocabulary, never connect to the internet, and only act when somebody is addressing you. Which, in turn, gets rid of the complicated set-up and addressing interaction design issues of centralised voice assistants. No more “front room lights: lamp 1 turn on” because… you just look at it.
- A Practical Guide to Maintaining Machine Learning — As Mike Loukides says, “ops is unprepared for ML”. [S]ome practices I’ve found useful to maintaining machine learning in production.
- Measuring the Predictability of Life Outcomes with a Scientific Mass Collaboration — Hundreds of researchers attempted to predict six life outcomes, such as a child’s grade point average and whether a family would be evicted from their home. These researchers used machine-learning methods optimized for prediction, and they drew on a vast dataset that was painstakingly collected by social scientists over 15 y. However, no one made very accurate predictions. For policymakers considering using predictive models in settings such as criminal justice and child-protective services, these results raise a number of concerns. Additionally, researchers must reconcile the idea that they understand life trajectories with the fact that none of the predictions were very accurate.