Showing posts with the label criminal justice

How I'm Voting in California's Overwhelming June Primaries

I've finished going through the June 5th California primary election for Oakland, and below are my picks. If anyone disagrees with my choices, please make your case—most of these I do not feel strongly about, and I did not have time to thoroughly research all of them.

In general, the issue I care most about is opposing or at least not supporting animal agriculture. In most cases that's irrelevant, as I could not find any relevant positions. The rare exception is incumbent Senator Dianne Feinstein, who spearheaded the Animal Enterprise Terrorism Act, which is disqualifying. On top of that, she's historically supported mass incarceration and has buckled under to Trump until she started fearing primary challengers. Senator Feinstein deserves to lose badly. Everyone should vote for her most plausible opponent from the left, Kevin DeLeon, both now and in November. On top of that, if you can state publicly that you voted against her because of the AETA, that would be powerful.

As …

Data and Racism in Machine Learning?

We often hear stories these days about racism in machine learning algorithms. The subtlety in these stories is often missing. I've been reading about this recently and found this quote very telling:
A wave of scholarship, triggered by the ProPublica report, illuminated the statistical challenge at the heart of the argument: Given that the underlying “base rate” of rearrest is higher for blacks than for whites, it is mathematically inevitable that the burden of false positives will fall more heavily on black defendants than on white ones. In other words, given that more black defendants than white defendants actually do have a high risk of reoffending, a “high risk” label that is correct 70% of the time for both white and black defendants will still mis-label more black than white defendants as high risk. A study titled “Inherent Tradeoffs in the Fair Determination of Risk Scores” proved mathematically that when rearrest rates are not equal between races, a well-calibrated tool like…