How to Upgrade Judges with Machine Learning

cageymaru

Fully [H]
Joined
Apr 10, 2003
Messages
22,086
Judges have often been criticized for locking up nonviolent offenders while awaiting trial and missing clues on potential violent offenders that hurt more citizens when allowed back onto the street. Sometimes issues of race, economic status, and more come into play at sentencing or a parole hearing. This means that more offenders are held than need be which clogs the wheels of justice. The National Bureau of Economic Research in conjunction with scientists and economists have devised an algorithm to assist judges at doing their job.

To accomplish this task, National Bureau of Economic Research studied over 100k judicial cases paying close attention to the offense they are suspected of, when and where they were arrested, and numbers and type of prior convictions. Age was used as a demographic marker but not race. When applied to 100k more cases, the results of the machine learning algorithm were better than the judges in determining flight risk and propensity to commit another crime. Then applying the algorithm to court cases in 40 urban cities yielded the same results.

In New York City alone they determined that they could cut crime by offenders awaiting trial by 25% if leaving the same amount of offenders jailed. Or they could lower the prison population by 40% while leaving the repeat offender crime rate the same. The system could be a fail safe type system where it allows the judge to make all decisions and the software will flag cases where the machine learning believes the judge is wrong. Some judges miss behavioral markers that a person is a flight risk or repeat offender.

What do you think? This seems like it could be a great tool to assist with the court system as 40% less prisoners to feed and guard sounds awesome. By the same token other machine learning software packages were deemed too harsh on certain races as they used different markers to determine potential for recidivism. Is this the beginning of a Minority Report system?

Kleinberg suggests that algorithms could be deployed to help judges without major disruption to the way they currently work in the form of a warning system that flags decisions highly likely to be wrong. Analysis of judges’ performance suggested they have a tendency to occasionally release people who are very likely to fail to show in court, or to commit crime while awaiting trial. An algorithm could catch many of those cases, says Kleinberg.
 
I think that if the tin-foil hats are correct... a computer saying "You're wrong, don't do it." won't change a thing.
 
Working in the NYS Criminal court system I can tell you not a single Judge will listen to a machine telling them they made a wrong decision.

Pick whatever wording you want - Machine Checking, Machine Support, etc. The Judges will only perceive this as them being told how to do their job. And if there's one thing a Judge NEVER wants to hear, it's this.
 
This will disproportionately affect minorities and the poor. I am not sure how but I know it will.
 
They disproportionately commit the crimes. So of course it will. But can an algorithm be classist or racist?
 
This is a great start, and I'd love to see it in action. Another would be to actually make the prison system rehabilitative instead of punitive. Problem is, there's just too much money in it, and the prison lobby is too powerful.
 
They disproportionately commit the crimes. So of course it will. But can an algorithm be classist or racist?

I was being sarcastic. That being said I think the algorithm inherently is not but the programmer can be. So you could bias the algorithm through the programmer...
 
They disproportionately commit the crimes. So of course it will. But can an algorithm be classist or racist?

All walks of people commit crimes. I agree that the poor are more likely to commit a crime due to the perils of that situation alone. In most societies minorities tend to comprise a large percentage of the poor, but you will find that the poor amongst the majority commit the same crimes at the same rate.

In the end the common denominator is being poor.
 
All walks of people commit crimes. I agree that the poor are more likely to commit a crime due to the perils of that situation alone. In most societies minorities tend to comprise a large percentage of the poor, but you will find that the poor amongst the majority commit the same crimes at the same rate.

In the end the common denominator is being poor.

Yup, and minorities being repressed due to their "perceived affiliation with crime" makes them increasingly poor, its a vicious cycle.
 
Just one more step in the impending robotic takeover.

Next we'll put this AI into a Cyborg, and it will be judge, jury, and executioner on the streets of Detroit. But alas, the antihero not be thugs and gansters, it will be a full robotic version, much larger, gone rogue. The compassion and emotion of the cyborg's human side will shine though and save the day.
 
Many of the weird sentences are due to stupid laws passed years ago in response to "fill in the blank" crisis that resulted in some kind of knee jerk legislative action.

Oklahoma example. Illegal Invader in an unregistered car driving without a license performed an illegal u-turn and killed a guy on a scooter. Max charge available to prosecutors for that offense carried a 5 year sentence. During the police search of the subject, they found tucked in his wallet a small packet of cocaine. Drugs were not a factor in the accident according to tests done. Yet due to anti-drug laws, Prosecutors were able to charge possession which could result in a 10 year sentence. Subject was found guilty on both charges. Result was concurrent sentences which had the net effect of zero punishment for killing someone due to a series of illegal actions but a 10 year sentence for simply possessing a small personal use size packet of cocaine.

Fix the weird laws first and judges could do a better job on sentencing. We passed such a reform in OK last November but a few legislators are convinced we voters were unable to understand the issue and are trying to pass a bill to undue the reforms.
 
All walks of people commit crimes. I agree that the poor are more likely to commit a crime due to the perils of that situation alone. In most societies minorities tend to comprise a large percentage of the poor, but you will find that the poor amongst the majority commit the same crimes at the same rate.

In the end the common denominator is being poor.

Poor is a vicious thing. People living in very bad areas, raising their kids in the environment, next generation the same thing. BUT, morals are morals and right from wrong are the same among the rich and poor. So no excuse for breaking the law. Can't use wealth as an excuse.
 
Working in the NYS Criminal court system I can tell you not a single Judge will listen to a machine telling them they made a wrong decision.

Pick whatever wording you want - Machine Checking, Machine Support, etc. The Judges will only perceive this as them being told how to do their job. And if there's one thing a Judge NEVER wants to hear, it's this.

Those judges will be valuable new data points that, if the statistics of the article hold true, lead to more support for incorporating machine judgement. Judges who have access to the machine provided data but make contrary judgement found to induce unnecessary cost to the state, would be visible.
 
Poor is a vicious thing. People living in very bad areas, raising their kids in the environment, next generation the same thing. BUT, morals are morals and right from wrong are the same among the rich and poor. So no excuse for breaking the law. Can't use wealth as an excuse.

Oh I wasn't giving them a pass for being poor. I bet 95% of people that grow up poor don't commit crimes. By the same token I have seen some truly poor souls steal food for their children because they couldn't afford a residence and thus the necessary address to get food stamps. The local cops bought them food and none of it was wasted! Lady ended up getting a job because of the cops and I bet she's still working today.

I was just saying that being poor seems to be the common denominator. I agree with you 100%.
 
Oh I wasn't giving them a pass for being poor. I bet 95% of people that grow up poor don't commit crimes. By the same token I have seen some truly poor souls steal food for their children because they couldn't afford a residence and thus the necessary address to get food stamps. The local cops bought them food and none of it was wasted! Lady ended up getting a job because of the cops and I bet she's still working today.

I was just saying that being poor seems to be the common denominator. I agree with you 100%.

And don't think for a minute I don't feel for their plight. I hate it for them and the country. Yeah, we are on the same boat here with our thinking.
 
Could work great. But too often I see "machine learning" used in the same way that "sprinkle fairy dust" is.

"This isn't possible. Sure it is, because, um...machine learning?"
 
Could work great. But too often I see "machine learning" used in the same way that "sprinkle fairy dust" is.

"This isn't possible. Sure it is, because, um...machine learning?"

When you get old and deemed useless, they will "machine learn" you into a new job as a host. Muhahaha!
 
This is a great start, and I'd love to see it in action. Another would be to actually make the prison system rehabilitative instead of punitive. Problem is, there's just too much money in it, and the prison lobby is too powerful.

I disagree, the rehad proponents had their shot and it's not only a failed policy it's expensive as hell. The only thing inmates should be learning is how to make little rocks from big rocks.

Longer sentences just give them time to start with bigger rocks.
 
Could work great. But too often I see "machine learning" used in the same way that "sprinkle fairy dust" is.

"This isn't possible. Sure it is, because, um...machine learning?"
So true. My company is working on a large healthcare quality/safety predictive analytics project so we talk to a lot of healthcare tech companies. Every other one is marketeers going "machine learning! AI!" and they all wilt when questioned. I think 3 companies in all of healthcare actually do the real deal.
 
I disagree, the rehad proponents had their shot and it's not only a failed policy it's expensive as hell. The only thing inmates should be learning is how to make little rocks from big rocks.

Longer sentences just give them time to start with bigger rocks.

I'm not sure what you mean by "had their shot," as the prison system in America has long been used punitively, in segregation and racial suppression. We've never had a prison system in America that is actually geared for rehabilitation. Approaches that treat convicts humanely lead to more wholesome societies, with tiny prison populations (and costs) and more productive after-prison lives for those who have been committed (which leads to a higher per capita GDP). This approach can be observed in many countries, including but not limited to Australia, Singapore, Norway, and Finland.

Regardless of political leaning, increasing the criminal stigma and societal divide is expensive for the country and generally leads to more crime. I personally want less crime, more wealth, and at lower costs.
 
I'm not sure what you mean by "had their shot," as the prison system in America has long been used punitively, in segregation and racial suppression. We've never had a prison system in America that is actually geared for rehabilitation. Approaches that treat convicts humanely lead to more wholesome societies, with tiny prison populations (and costs) and more productive after-prison lives for those who have been committed (which leads to a higher per capita GDP). This approach can be observed in many countries, including but not limited to Australia, Singapore, Norway, and Finland.

Regardless of political leaning, increasing the criminal stigma and societal divide is expensive for the country and generally leads to more crime. I personally want less crime, more wealth, and at lower costs.

There was a time when US prisons were seriously punitive in nature and the prison populations were much smaller, even during trying time like prohibition. In the 60's we started seriously working toward prison reform and initially that reform targeted the "human issues" and problems with prison corruption, the movie Brubaker comes to mind. But the reform drive didn't stop there and from the 70s through the 90s we continued along the lines of rehabilitation. No inmate ever came out of a prison with a law degree prior to this era, the reform you are talking about has come, failed, and now we are just living with a system that is returning to it's punitive roots while it drags the baggage of the reform effort along for the ride. Failed garbage programs and costs and procedures, going through the motions. Reform failed because in the most simple terms, the drug trade is more lucrative and all it takes is willingness to join up and bare the risks of failure.

Rehabilitation failed and no one can say it was never given a fair shot. It just couldn't compete.
 
There was a time when US prisons were seriously punitive in nature and the prison populations were much smaller, even during trying time like prohibition. In the 60's we started seriously working toward prison reform and initially that reform targeted the "human issues" and problems with prison corruption, the movie Brubaker comes to mind. But the reform drive didn't stop there and from the 70s through the 90s we continued along the lines of rehabilitation. No inmate ever came out of a prison with a law degree prior to this era, the reform you are talking about has come, failed, and now we are just living with a system that is returning to it's punitive roots while it drags the baggage of the reform effort along for the ride. Failed garbage programs and costs and procedures, going through the motions. Reform failed because in the most simple terms, the drug trade is more lucrative and all it takes is willingness to join up and bare the risks of failure.

Rehabilitation failed and no one can say it was never given a fair shot. It just couldn't compete.

That sounds most like what I'm getting at, in that it shouldn't need to compete. I recognize the prior reformative efforts, and that a 30% reform rate is better than a 0% reform rate. I merely think we could do better; but, we'd have to first separate prisons from profit, and maybe halt lobbying altogether. May not happen without revolution.
 
That sounds most like what I'm getting at, in that it shouldn't need to compete. I recognize the prior reformative efforts, and that a 30% reform rate is better than a 0% reform rate. I merely think we could do better; but, we'd have to first separate prisons from profit, and maybe halt lobbying altogether. May not happen without revolution.
I get you.

I'm not sure about the Prisons for Profit angle, I just don't know enough and on the surface of it I don't see the connection between the people sending people to prison and the one's profiting from it.
 
Last edited:
I get you.

I'm not sure about the Prisons for Profit angel, I just don't know enough and on the surface of it I don't see the connection between the people sending people to prison and the one's profiting from it.

I recognize you didn't ask, and won't blame you if you choose not to investigate. This is simply something I'm vociferously passionate about. There's a pretty strong connection; I referenced lobbying for a reason.
https://www.washingtonpost.com/post...ome-the-biggest-lobby-no-one-is-talking-about

It gets a bit more insidious if the racial context is added.
https://www.researchgate.net/public..._Reassess_State_Takeover_District_Performance
 
I recognize you didn't ask, and won't blame you if you choose not to investigate. This is simply something I'm vociferously passionate about. There's a pretty strong connection; I referenced lobbying for a reason.
https://www.washingtonpost.com/post...ome-the-biggest-lobby-no-one-is-talking-about

It gets a bit more insidious if the racial context is added.
https://www.researchgate.net/public..._Reassess_State_Takeover_District_Performance


Ummm, I was reading the first article linked and I reached this quote and I must admit, I lost all interest.
They now rake in a combined $3.3 billion in annual revenue and the private federal prison population more than doubled between 2000 and 2010, according to a report by the Justice Policy Institute.

I know 3.3 billion sounds like a large number, but it is in fact, chump change.

I thought I knew what "a lot of money was" until in 2003, I read an article that the IRS was going to make Symantec pay an additional 1 Billion in taxes because they said Symantec had under-valued an intellectual property transfer from one of their overseas offices.

I read this, and I tried to put it into perspective. An additional 1 billion in taxes owed ....... How many Billions in taxes does a decent sized company generate from how many Billions in profits ?

How many companies out there make this kind of money, Symantec isn't a company I think of as large really, IBM, Microsoft, Google, yes, Symantec ?

I realized that I can not actually conceive of how much money is out there in the world. It's like asking someone to conceptualize how large outer space is, you know, all of it.

And this was 2003, 14 years ago.

$3.3 Billion sounds like a lot of money to you and I when we don't think too hard about it. But from the viewpoint of big business, it's peanuts ... chump change. The article is hype;

The two largest for-profit prison companies in the United States – GEO and Corrections Corporation of America – and their associates have funneled more than $10 million to candidates since 1989

Since 1989, that's what? almost 30 years, 10 million divided by 28 years is $357,000+ a year. It's the price of a decent sized house in a moderate cost of living region. If you divided that among more then a few candidate's campaigns ......... It no longer has the impact does it?

I'm sorry, these articles do not paint a picture that looks troubling for me.

There is a quote from CCA about how lessening punishments or decriminalizing something like say, pot, could have a negative impact on their business. CCA is a business, this is just common sense normal concerns for any business. I understand that it sounds bad, that they would lobby to keep decriminalization in place and prison pops stable or growing. But any business would. The only difference is what their business is.

I can see there is an issue. And I am sure it probably effects some people. I am not so sure, given my casual look into it, that this is anywhere near the problem the writer tries to make it sound like.

Prison is a nasty business, it's a nasty business even when it's not a business. If private for-profit prisons are not cost effective for the State and Federal governments then I can see a real issue. But if they are cheaper for the tax payer then there is less traction to argue against them.
 
Last edited:
They are cheaper, except, prison as a business has a vested interest in keeping the cells full. One owned by the government does not. In that situation, you have to wonder how many are still there simply to keep the cells full, while the private prisons lobby for more money for more prisons cause this one is full up. I am all for letting the private sector handle most things, law enforcement and criminal facilities is not one of them.

Legalize most drugs, tax the shit out of them, let every convict in prison on drug charges not involving violence, or trafficking from other countries go. Give those recently freed individuals the jobs of the illegals Trump is going to try and get rid of. Profit?

That was somewhat tongue in cheek, but something similar to that would likely be to the benefit of our country.
 
They are cheaper, except, prison as a business has a vested interest in keeping the cells full. One owned by the government does not. In that situation, you have to wonder how many are still there simply to keep the cells full, while the private prisons lobby for more money for more prisons cause this one is full up. I am all for letting the private sector handle most things, law enforcement and criminal facilities is not one of them.

Legalize most drugs, tax the shit out of them, let every convict in prison on drug charges not involving violence, or trafficking from other countries go. Give those recently freed individuals the jobs of the illegals Trump is going to try and get rid of. Profit?

That was somewhat tongue in cheek, but something similar to that would likely be to the benefit of our country.


I understand your concern. But if the prison doesn't have any say-so in who goes to prison or who stays in prison, then I am not sure I see a serious issue. And as I just showed, the lavish lobbying as reported is actually minuscule and the profits are actually not all that big at all. When I see reporting and claims made claiming a serious problem exists and then I find that some of the claims look greatly and unreasonably exaggerated, it casts doubt on the rest. Even if the rest has substance, it creates doubt.
 
They do have a say in who stays. They have control of an inmate, they can easily increase the amount of time a person stays in jail. It is relatively easy to tack time on over contraband, or fighting, or other criminal actions. Which is pretty easy to frame up. They also decide what constitutes good/bad behavior. So they have a say in the parole proceedings as well.

Of course it would help if criminals would all stop committing crimes while in prison. It is ridiculous just how much crime goes on inside those walls. However, because it goes on, it is something that can be manipulated as needed by the prison, since it can easily go unnoticed. It comes down to a matter of trust. Do I trust corporations with a financial interest in keeping an inmate in prison to do the correct/just thing? I can't find it in me to trust them under such a situation. A person can sometimes be trusted to do what is right even when it hurts them, corporations, as a rule can not.
 
Back
Top