Search
Thomas Dietterich
675 followers
675 followers
Thomas's posts
Post has shared content
Neil Lawrence says many important things in this blog post. Very insightful!
A new blog post on "What Kind of AI have we Created". This is a theme I've been developing in a few talks recently.
http://inverseprobability.com/2015/12/04/what-kind-of-ai/
http://inverseprobability.com/2015/12/04/what-kind-of-ai/
Is the big red bar in the new G+ interface just for the holidays, or will we need to look at it all year round? #G+
Post has shared content
Things are getting very interesting at BigML
Post has attachment
Spam journal spam!
I just received this email:
"Start you own journal from your country. You will get some publishing charges in dollars. Will you act as an Editor-in-Chief for some New International Research Journals? If yes, we are happy to inform you to fill the following form.
[...]
*M.K.R.S. VEERA KUMAR will do the web hosting and other technical works to publish the journal online.
Start your own journal from your country.
Sincerely Yours,
Journal Development Member
"
I just received this email:
"Start you own journal from your country. You will get some publishing charges in dollars. Will you act as an Editor-in-Chief for some New International Research Journals? If yes, we are happy to inform you to fill the following form.
[...]
*M.K.R.S. VEERA KUMAR will do the web hosting and other technical works to publish the journal online.
Start your own journal from your country.
Sincerely Yours,
Journal Development Member
"
Post has shared content
This was a fascinating story!
Post has shared content
We are hiring for a joint position between CS and robotics. Potential areas include all of the usual areas, but also robot vision and robot prosthetics.
Robotics friends: We are hiring new faculty in Robotics, please let people know! People with questions are welcome to Email me.
http://robotics.oregonstate.edu/jobs
We're just moving into our new space, and it's pretty nice... It is a good time to be joining, as the University is really waking up to the potential of Robotics and supporting our program.
http://robotics.oregonstate.edu/jobs
We're just moving into our new space, and it's pretty nice... It is a good time to be joining, as the University is really waking up to the potential of Robotics and supporting our program.
Post has shared content
Amazing data analysis error!
This is one of those news items that hasn't gotten nearly enough coverage -- because it's the sort of thing that makes professionals go OH YOU HAVE GOT TO BE FUCKING KIDDING ME.
What happened? Back in 2005, the Bureau of Justice Statistics (a branch of the DOJ) did a study on recidivism, and found out that the rate is tremendously high: 68% of state prisoners end up back behind bars within three years of release. Once a criminal, always a criminal, they concluded -- and people have been shaping policy to match.
But a team read through it carefully, and it turns out that the BJS made a basic, bonehead, mistake in their statistical analysis. They thought they were measuring whether people who go to prison will reoffend; what they actually measured was that most people in prison, on any given day, are repeat offenders.
Which makes sense, because repeat offenders spend a lot more time in prison than one-time offenders.
These are not the same thing. At all. It turns out that if you do the analysis right, only 30% or so of prisoners will ever re-offend, and only 11% will do so multiple times. In fact, this "once a criminal, always a criminal" rule appears to be completely false -- unless, that is, you structure policies so that anyone with a criminal conviction is treated like a permanent criminal, and so not allowed to (say) get virtually any job other than "criminal." In which case, you will in fact end up with lots of criminals.
In the post linked below, +Andreas Schou gives some of the explanation of what went wrong in the study. You can read more at the linked Slate article (http://www.slate.com/articles/news_and_politics/crime/2015/10/why_do_so_many_prisoners_end_up_back_in_prison_a_new_study_says_maybe_they.html), and even more with the paper that actually found the mistake. (http://cad.sagepub.com/content/early/2014/09/26/0011128714549655.abstract)
The most important lesson in all of this is that it's easy to make bonehead mistakes in statistics. If the statistics matter -- if you're going to use them to prescribe drugs or set public policy or something like that -- it's very important to have people check your work, repeatedly, and ask the right questions. The most important question is "have you actually measured what you think you measured," because there are all sorts of ways to screw that up.
There's also a great new book on that subject: Alex Reinhart's Statistics Done Wrong. (http://www.statisticsdonewrong.com/) Please, if you do statistics in your daily life, read it.
What happened? Back in 2005, the Bureau of Justice Statistics (a branch of the DOJ) did a study on recidivism, and found out that the rate is tremendously high: 68% of state prisoners end up back behind bars within three years of release. Once a criminal, always a criminal, they concluded -- and people have been shaping policy to match.
But a team read through it carefully, and it turns out that the BJS made a basic, bonehead, mistake in their statistical analysis. They thought they were measuring whether people who go to prison will reoffend; what they actually measured was that most people in prison, on any given day, are repeat offenders.
Which makes sense, because repeat offenders spend a lot more time in prison than one-time offenders.
These are not the same thing. At all. It turns out that if you do the analysis right, only 30% or so of prisoners will ever re-offend, and only 11% will do so multiple times. In fact, this "once a criminal, always a criminal" rule appears to be completely false -- unless, that is, you structure policies so that anyone with a criminal conviction is treated like a permanent criminal, and so not allowed to (say) get virtually any job other than "criminal." In which case, you will in fact end up with lots of criminals.
In the post linked below, +Andreas Schou gives some of the explanation of what went wrong in the study. You can read more at the linked Slate article (http://www.slate.com/articles/news_and_politics/crime/2015/10/why_do_so_many_prisoners_end_up_back_in_prison_a_new_study_says_maybe_they.html), and even more with the paper that actually found the mistake. (http://cad.sagepub.com/content/early/2014/09/26/0011128714549655.abstract)
The most important lesson in all of this is that it's easy to make bonehead mistakes in statistics. If the statistics matter -- if you're going to use them to prescribe drugs or set public policy or something like that -- it's very important to have people check your work, repeatedly, and ask the right questions. The most important question is "have you actually measured what you think you measured," because there are all sorts of ways to screw that up.
There's also a great new book on that subject: Alex Reinhart's Statistics Done Wrong. (http://www.statisticsdonewrong.com/) Please, if you do statistics in your daily life, read it.
Given the amount of click bait on G+, FB, Twitter, etc. I would like to have a reading app that would examine the linked item and give me a gist so that I could decide whether I really want to click.
Alternatively, can someone cook up a "Click Bait Block" app, kind of like AdBlockPlus, so that people need to give more information to get around the blocker?
Alternatively, can someone cook up a "Click Bait Block" app, kind of like AdBlockPlus, so that people need to give more information to get around the blocker?
Post has attachment
Here is the CACM Perspective that Eric Horvitz and I have written.
"The computer science community must take a leadership role in exploring and addressing concerns about machine intelligence. We must work to ensure that AI systems responsible for high-stakes decisions will behave safely and properly, and we must also examine and respond to concerns about potential transformational influences of AI. Beyond scholarly studies, computer scientists need to maintain an open, two-way channel for communicating with the public about opportunities, concerns, remedies, and realities of AI." #AI #CACM
"The computer science community must take a leadership role in exploring and addressing concerns about machine intelligence. We must work to ensure that AI systems responsible for high-stakes decisions will behave safely and properly, and we must also examine and respond to concerns about potential transformational influences of AI. Beyond scholarly studies, computer scientists need to maintain an open, two-way channel for communicating with the public about opportunities, concerns, remedies, and realities of AI." #AI #CACM
Wait while more posts are being loaded