From awd at cs.cmu.edu  Fri Jun  1 13:23:59 2018
From: awd at cs.cmu.edu (Artur Dubrawski)
Date: Fri, 1 Jun 2018 13:23:59 -0400
Subject: Fwd: PhD Speaking Qualifier: Design with Interpretability in Mind: An
 Alternate Ethos for Data Science
In-Reply-To: <CAJ4FYPy21pT7dNtN-7FomeZpzpqBR0v8eHZiMhEpk8FE+epW-Q@mail.gmail.com>
References: <CAJ4FYPy21pT7dNtN-7FomeZpzpqBR0v8eHZiMhEpk8FE+epW-Q@mail.gmail.com>
Message-ID: <CAJvAoyvzUHPX91Hi+KDg8WONTQoLHNGgnVCAUwTVOT+4oJEWfQ@mail.gmail.com>

fyi, a good talk to attend!

---------- Forwarded message ----------
From: Nick Gisolfi <ngisolfi at cs.cmu.edu>
Date: Fri, Jun 1, 2018 at 1:15 PM
Subject: PhD Speaking Qualifier: Design with Interpretability in Mind: An
Alternate Ethos for Data Science
To: ri-people at cs.cmu.edu, Artur Dubrawski <awd at cs.cmu.edu>, Deva Kannan
Ramanan <deva at cs.cmu.edu>, Barnabas Poczos <bapoczos at cs.cmu.edu>, Matt
Barnes <mbarnes1 at andrew.cmu.edu>


Hi Everyone,


I will be giving my Ph.D. speaking qualifier this Monday, June 4th at 3pm
in GHC 8102.  I hope to see you there!


*Details*:

*Date*: Monday, June 4th

*Time*: 3-4pm

*Location*: GHC 8102


*Title*: Design with Interpretability in Mind: An Alternate Ethos for Data
Science


*Abstract*:

The fields of Machine Learning and Data Science generally follow the
paradigm that ?the ends justify the means?, where improving predictive
power of an algorithm is considered of paramount value, even when
implemented at the expense of model intelligibility.  While accuracy is an
important performance metric, interpretability should be a major
consideration for many application domains.  This is particularly true for
decision support systems where a human must ultimately take responsibility
for their decision based on machine recommendations.  Other times, applying
the most powerful state-of-the-art learning models to data may not be
necessary in order to make confident predictions, and in those cases,
interpretability can be maintained by keeping algorithms as simple as
possible.


I will be sharing a novel bounding box algorithm which finds
easy-to-understand, low-dimensional structure in data.  I will then discuss
a few use cases where we can leverage these simple structures to provide
interpretable answers to potentially complex questions.  Finally, I will
show that for a given data set, some data are ?harder? than others.  I will
present a staged model framework which provides interpretable predictions
for the ?easy? data, while allowing the ?hard? data to be processed by a
more powerful and more complex alternative model.  Across a survey of some
publicly available data sets, I will show that a significant amount of data
can be confidently handled with a simple model, without incurring a
statistically distinguishable loss in accuracy compared to a more powerful
black-box model.



*Committee*:

Artur Dubrawski

Barnab?s P?czos

Deva Ramanan

Matt Barnes


Thanks!


- Nick
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20180601/249f9801/attachment.html>

From vjeansel at andrew.cmu.edu  Mon Jun  4 08:37:46 2018
From: vjeansel at andrew.cmu.edu (Vincent Jeanselme)
Date: Mon, 4 Jun 2018 08:37:46 -0400
Subject: Scratch Use - Lov 3
Message-ID: <b59c22e6-ed73-5200-4a4a-133649bd440c@andrew.cmu.edu>

Good Morning All,

Could you check your disk usage on the scratch directory of lov3 and 
clean it if necessary : 100% of the disk is used.

Thank you,

-- 
Vincent Jeanselme


From awd at cs.cmu.edu  Tue Jun  5 15:59:28 2018
From: awd at cs.cmu.edu (Artur Dubrawski)
Date: Tue, 5 Jun 2018 15:59:28 -0400
Subject: Auton Lab guest lecture this Friday: Mathieu Guillame-Bert
Message-ID: <CAJvAoyvVjK=iaBpNyjyGK9jnRMMPr2ftCGfqssLmu6MPmH27cQ@mail.gmail.com>

 Dear Autonians,

On Friday, we can all witness a triumphant return of our own Dr.
Guillame-Bert
who will be visiting Pittsburgh for  few days.

He has graciously agreed to give us a talk, see below for its title and
abstract.

It will take place in NSH 3305 and Jessie and Predrag will decide where to
order lunch.
Please confirm with them (cc-d here for convenience) if you're planning to
attend
so that we can get an approximate head count to order food.

Cheers!
Artur

TITLE:
Applying Random Forest at Google

ABSTRACT:
Random Forest is a popular supervised learning algorithm that shows both
high usability and high performances. For this reason, Random Forest is
often a competitive candidate solution for machine learning tasks in
industrial environments. However, the amount of training data and the
serving constraints present in Google, gives rise to many challenges in its
use.
In this talk, I will present recent developments of Random Forest
technology for Google usage, including 1) The distributed training of
Random Forest over hundreds of computers in order to handle datasets with
billions of examples. 2) The efficient tuning of Random Forest
hyper-parameters to any custom objective function with constraints. 3) And
finally, early work on using GPUs to speeding-up the inference of Random
Forest models.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20180605/407889b5/attachment.html>

From awd at cs.cmu.edu  Tue Jun  5 21:34:40 2018
From: awd at cs.cmu.edu (Artur Dubrawski)
Date: Tue, 5 Jun 2018 21:34:40 -0400
Subject: NSH 3305 at NOON [Re: Auton Lab guest lecture this Friday: Mathieu
 Guillame-Bert]
Message-ID: <CAJvAoyugpSRThcpJ8Wd5sk6FkW__XNCboJQwOZ6ryf_TQ8FO0Q@mail.gmail.com>

As Jarod astutely pointed out, the time was missing in the original
announcement. It will be high noon.


On Tue, Jun 5, 2018 at 3:59 PM, Artur Dubrawski <awd at cs.cmu.edu> wrote:

> Dear Autonians,
>
> On Friday, we can all witness a triumphant return of our own Dr.
> Guillame-Bert
> who will be visiting Pittsburgh for  few days.
>
> He has graciously agreed to give us a talk, see below for its title and
> abstract.
>
> It will take place in NSH 3305 and Jessie and Predrag will decide where to
> order lunch.
> Please confirm with them (cc-d here for convenience) if you're planning to
> attend
> so that we can get an approximate head count to order food.
>
> Cheers!
> Artur
>
> TITLE:
> Applying Random Forest at Google
>
> ABSTRACT:
> Random Forest is a popular supervised learning algorithm that shows both
> high usability and high performances. For this reason, Random Forest is
> often a competitive candidate solution for machine learning tasks in
> industrial environments. However, the amount of training data and the
> serving constraints present in Google, gives rise to many challenges in its
> use.
> In this talk, I will present recent developments of Random Forest
> technology for Google usage, including 1) The distributed training of
> Random Forest over hundreds of computers in order to handle datasets with
> billions of examples. 2) The efficient tuning of Random Forest
> hyper-parameters to any custom objective function with constraints. 3) And
> finally, early work on using GPUs to speeding-up the inference of Random
> Forest models.
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20180605/a8946cd2/attachment-0001.html>

From chiragn at cs.cmu.edu  Wed Jun  6 10:38:48 2018
From: chiragn at cs.cmu.edu (Chirag Nagpal)
Date: Wed, 6 Jun 2018 10:38:48 -0400
Subject: AAAI AI for Social Impact
Message-ID: <CAKH1gVrKsYtemdNv2KMBo88t3Mj8+KZ3OX7Z26QTjt0WciDOKg@mail.gmail.com>

AAAI has a new track this year, 'AI for Social Impact'. Submissions will
follow the same style and be presented in the main conference.

https://aaai.org/Conferences/AAAI-19/aaai19emergingcall/


This might be a highly relevant for some of our more applied work, and so I
thought i'll share


Chirag

-- 

*Chirag Nagpal* Graduate Student, Language Technologies Institute
School of Computer Science
Carnegie Mellon University
cs.cmu.edu/~chiragn
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20180606/f018eeef/attachment.html>

From awd at cs.cmu.edu  Thu Jun  7 12:50:13 2018
From: awd at cs.cmu.edu (Artur Dubrawski)
Date: Thu, 7 Jun 2018 12:50:13 -0400
Subject: Fwd: Sex Trafficking Victim Found Online from 2 Year Old Photo
In-Reply-To: <431757831d2c1c5ec10ecfdf9.edcf27e3bc.20180607162316.27d4e03631.f376f554@mail71.atl51.rsgsv.net>
References: <431757831d2c1c5ec10ecfdf9.edcf27e3bc.20180607162316.27d4e03631.f376f554@mail71.atl51.rsgsv.net>
Message-ID: <CAJvAoysZrvB38B_rPZu4eKc1HNvDVKBDiJ+6VvvQHszLv=U+=g@mail.gmail.com>

Sometimes really good things come out of our work.

Congrats to the Traffic Jam team!

Cheers,
Artur

---------- Forwarded message ---------
From: Marinus Analytics <info at marinusanalytics.com>
Date: Thu, Jun 7, 2018 at 12:39 PM
Subject: Sex Trafficking Victim Found Online from 2 Year Old Photo
To: <awd at cs.cmu.edu>


It started with an outcry from Allie, who was pimped 2 years ago by a
violent trafficker who went by the name ?Julian.?
<https://marinusanalytics.us3.list-manage.com/track/click?u=431757831d2c1c5ec10ecfdf9&id=8cc594f33b&e=edcf27e3bc>
Traffic Jam Success Story

?Justice is something for which every generation has to strive.?
? John M. Perkins

*The Outcry*
It started with an outcry from Allie,* who was pimped 2 years ago by a
violent trafficker who went by the name ?Julian.?* She told Detective John
Patterson,* ?I want to get out of this, because of what Julian?s done to
me. And he did it to a 15 year old girl, too.? Similar cases of this
scope?which grew to 21 identified victims?would take a year; Detective
Patterson built this case in about 3 months.  Detective Patterson credits
the importance of good experience, trainings, and technology tools like
Traffic Jam.

*The Violent Pimp*
Julian was a violent pimp, and he required a $1,500 per day quota for each
victim; he would beat them if they didn?t bring home that money each night.
He had been arrested many times in the past; for aggravated assault with a
motor vehicle, running his victims over with a car several times. For
strangling one of his victims until she passed out. For punching and
assaulting others. But force, fraud, and coercion aren?t always physical;
he also threatened to kill their children if they didn?t work for him.

Julian recruited his victims in person and on social media apps like
Instagram. He broke down one victim by recruiting her to work as a
stripper, and then repeatedly raping her over a number of days. When she
still refused to sell sex for him, he withheld food until she broke down
and agreed. When she tried to escape, he used a location tracking app on
her phone to chase her down. He found her, assaulted her, and put her back
to work. She couldn?t work for a few more days, because customers would
notice her black eye.

*Traffic Jam*
Detective Patterson used Allie?s testimony to begin piecing together the
case. By searching victims? Facebook photos through Traffic Jam?s
FaceSearch, he was able to find their ads posted across the country, from
the South all the way up to Pennsylvania and back. He said, ?I used Traffic
Jam to map out the course that Allie exactly described.?

He was searching for Jessica,* an underage victim Allie had told him about.
He scrolled through Jessica?s Instagram, and found the most recent pictures
she posted of herself, which were over 2 years old, from when she was
fifteen. He said, ?I didn?t think it would lead to anything, because it was
such an old photo. But I thought I?d run it through FaceSearch just in
case. I couldn?t believe it when the 2 year old photo returned top matches
in FaceSearch that looked just like her.? He saw in the Traffic Jam trail
that she was currently posting in California, but had posted in his city in
the months before. Through use of FaceSearch and confirmation of the
locations/times in the Traffic Jam trail, Detective Patterson successfully
identified Jessica using a photo that was 2 years old.

Allie told him about another victim, Sammy,* and her Facebook account. He
found some year-old pictures of Sammy on her Facebook profile and uploaded
them into FaceSearch to see if he could find a match. FaceSearch returned
top matches, none of which looked like Sammy. Patterson said, ?I thought
the matches weren?t her, they just didn?t look like her.? He sanity-checked
the top matches, by checking the timing and location of the ads. Then, he
said, ?I found that one of the phone numbers in the ad was registered to
her name. That made me realize that the pictures from the FaceSearch
results were a correct match, but I didn?t recognize her at first because
she had changed her appearance so drastically.? When the appearance of the
victim looked completely different, FaceSearch was still able to make a
positive match, in seconds.

*Where Are They Now*
By using various technology tools like Traffic Jam, paired with victim
interviews and evidence gathered through search warrants, Detective
Patterson determined the 21 victims Julian had exploited over a number of
years. Patterson assembled a history of money transfers from the victims to
their pimp. He confirmed that many of Julian?s victims were working in
different states (as correlated to their ads) and wiring their earnings
from out of state back to Julian. He determined Julian was making about
$15,000 a month, from 2 girls alone; and he had a total of 21 victims over
the span of the investigation.

The police department received an arrest warrant for Julian for 6 felonies,
which they served him when he came out of the house on trash day.
Detectives intervened right after he had threatened to kill another
victim?s baby if she didn?t keep working for him. This violent trafficker
is currently in jail without bail, awaiting prosecution and potential life
in prison for his crimes, all thanks to the tireless efforts of Detective
Patterson and his team.


* names have been changed to protect individual?s identity
*Contact*

For law enforcement who would like to be in contact with the detective
who submitted this story, please email us <info at marinusanalytics.com> and
we will connect you.

Have a success story other investigators should know about? Share your story
<info at marinusanalytics.com>.
<https://marinusanalytics.us3.list-manage.com/track/click?u=431757831d2c1c5ec10ecfdf9&id=64b84f5d80&e=edcf27e3bc>
<https://marinusanalytics.us3.list-manage.com/track/click?u=431757831d2c1c5ec10ecfdf9&id=5c04c2ea7f&e=edcf27e3bc>
<https://marinusanalytics.us3.list-manage.com/track/click?u=431757831d2c1c5ec10ecfdf9&id=141ae670dd&e=edcf27e3bc>
*Copyright ? 2018 *Marinus Analytics, All rights reserved.*



Want to change how you receive these emails?
You can update your preferences
<https://marinusanalytics.us3.list-manage.com/profile?u=431757831d2c1c5ec10ecfdf9&id=07193289f9&e=edcf27e3bc>
or unsubscribe from this list
<https://marinusanalytics.us3.list-manage.com/unsubscribe?u=431757831d2c1c5ec10ecfdf9&id=07193289f9&e=edcf27e3bc&c=27d4e03631>
.

[image: Email Marketing Powered by MailChimp]
<http://www.mailchimp.com/monkey-rewards/?utm_source=freemium_newsletter&utm_medium=email&utm_campaign=monkey_rewards&aid=431757831d2c1c5ec10ecfdf9&afl=1>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20180607/3190a1cd/attachment-0001.html>

From awm at cs.cmu.edu  Thu Jun  7 13:20:59 2018
From: awm at cs.cmu.edu (Andrew Moore)
Date: Thu, 7 Jun 2018 13:20:59 -0400
Subject: Sex Trafficking Victim Found Online from 2 Year Old Photo
In-Reply-To: <CAJvAoysZrvB38B_rPZu4eKc1HNvDVKBDiJ+6VvvQHszLv=U+=g@mail.gmail.com>
References: <431757831d2c1c5ec10ecfdf9.edcf27e3bc.20180607162316.27d4e03631.f376f554@mail71.atl51.rsgsv.net>
 <CAJvAoysZrvB38B_rPZu4eKc1HNvDVKBDiJ+6VvvQHszLv=U+=g@mail.gmail.com>
Message-ID: <CAOQLAA3qaksJ7cLMW4pAUis1cpz7y5C_vTci0Ld9=Jr83hwGsg@mail.gmail.com>

This is the perfect depiction of why I'm always so happy to get out of bed
in the morning. Warm congratulations.

Artur: this is public, so I assume it's okay to publicize this story?


On Thu, Jun 7, 2018 at 12:50 PM, Artur Dubrawski <awd at cs.cmu.edu> wrote:

> Sometimes really good things come out of our work.
>
> Congrats to the Traffic Jam team!
>
> Cheers,
> Artur
>
> ---------- Forwarded message ---------
> From: Marinus Analytics <info at marinusanalytics.com>
> Date: Thu, Jun 7, 2018 at 12:39 PM
> Subject: Sex Trafficking Victim Found Online from 2 Year Old Photo
> To: <awd at cs.cmu.edu>
>
>
> It started with an outcry from Allie, who was pimped 2 years ago by a
> violent trafficker who went by the name ?Julian.?
>
> <https://marinusanalytics.us3.list-manage.com/track/click?u=431757831d2c1c5ec10ecfdf9&id=8cc594f33b&e=edcf27e3bc>
> Traffic Jam Success Story
>
> ?Justice is something for which every generation has to strive.?
> ? John M. Perkins
>
> *The Outcry*
> It started with an outcry from Allie,* who was pimped 2 years ago by a
> violent trafficker who went by the name ?Julian.?* She told Detective John
> Patterson,* ?I want to get out of this, because of what Julian?s done to
> me. And he did it to a 15 year old girl, too.? Similar cases of this
> scope?which grew to 21 identified victims?would take a year; Detective
> Patterson built this case in about 3 months.  Detective Patterson credits
> the importance of good experience, trainings, and technology tools like
> Traffic Jam.
>
> *The Violent Pimp*
> Julian was a violent pimp, and he required a $1,500 per day quota for each
> victim; he would beat them if they didn?t bring home that money each night.
> He had been arrested many times in the past; for aggravated assault with a
> motor vehicle, running his victims over with a car several times. For
> strangling one of his victims until she passed out. For punching and
> assaulting others. But force, fraud, and coercion aren?t always physical;
> he also threatened to kill their children if they didn?t work for him.
>
> Julian recruited his victims in person and on social media apps like
> Instagram. He broke down one victim by recruiting her to work as a
> stripper, and then repeatedly raping her over a number of days. When she
> still refused to sell sex for him, he withheld food until she broke down
> and agreed. When she tried to escape, he used a location tracking app on
> her phone to chase her down. He found her, assaulted her, and put her back
> to work. She couldn?t work for a few more days, because customers would
> notice her black eye.
>
> *Traffic Jam*
> Detective Patterson used Allie?s testimony to begin piecing together the
> case. By searching victims? Facebook photos through Traffic Jam?s
> FaceSearch, he was able to find their ads posted across the country, from
> the South all the way up to Pennsylvania and back. He said, ?I used Traffic
> Jam to map out the course that Allie exactly described.?
>
> He was searching for Jessica,* an underage victim Allie had told him
> about. He scrolled through Jessica?s Instagram, and found the most recent
> pictures she posted of herself, which were over 2 years old, from when she
> was fifteen. He said, ?I didn?t think it would lead to anything, because it
> was such an old photo. But I thought I?d run it through FaceSearch just in
> case. I couldn?t believe it when the 2 year old photo returned top matches
> in FaceSearch that looked just like her.? He saw in the Traffic Jam trail
> that she was currently posting in California, but had posted in his city in
> the months before. Through use of FaceSearch and confirmation of the
> locations/times in the Traffic Jam trail, Detective Patterson successfully
> identified Jessica using a photo that was 2 years old.
>
> Allie told him about another victim, Sammy,* and her Facebook account. He
> found some year-old pictures of Sammy on her Facebook profile and uploaded
> them into FaceSearch to see if he could find a match. FaceSearch returned
> top matches, none of which looked like Sammy. Patterson said, ?I thought
> the matches weren?t her, they just didn?t look like her.? He sanity-checked
> the top matches, by checking the timing and location of the ads. Then, he
> said, ?I found that one of the phone numbers in the ad was registered to
> her name. That made me realize that the pictures from the FaceSearch
> results were a correct match, but I didn?t recognize her at first because
> she had changed her appearance so drastically.? When the appearance of the
> victim looked completely different, FaceSearch was still able to make a
> positive match, in seconds.
>
> *Where Are They Now*
> By using various technology tools like Traffic Jam, paired with victim
> interviews and evidence gathered through search warrants, Detective
> Patterson determined the 21 victims Julian had exploited over a number of
> years. Patterson assembled a history of money transfers from the victims to
> their pimp. He confirmed that many of Julian?s victims were working in
> different states (as correlated to their ads) and wiring their earnings
> from out of state back to Julian. He determined Julian was making about
> $15,000 a month, from 2 girls alone; and he had a total of 21 victims over
> the span of the investigation.
>
> The police department received an arrest warrant for Julian for 6
> felonies, which they served him when he came out of the house on trash day.
> Detectives intervened right after he had threatened to kill another
> victim?s baby if she didn?t keep working for him. This violent trafficker
> is currently in jail without bail, awaiting prosecution and potential life
> in prison for his crimes, all thanks to the tireless efforts of Detective
> Patterson and his team.
>
>
> * names have been changed to protect individual?s identity
> *Contact*
>
> For law enforcement who would like to be in contact with the detective
> who submitted this story, please email us <info at marinusanalytics.com> and
> we will connect you.
>
> Have a success story other investigators should know about? Share your
> story <info at marinusanalytics.com>.
>
> <https://marinusanalytics.us3.list-manage.com/track/click?u=431757831d2c1c5ec10ecfdf9&id=64b84f5d80&e=edcf27e3bc>
>
> <https://marinusanalytics.us3.list-manage.com/track/click?u=431757831d2c1c5ec10ecfdf9&id=5c04c2ea7f&e=edcf27e3bc>
>
> <https://marinusanalytics.us3.list-manage.com/track/click?u=431757831d2c1c5ec10ecfdf9&id=141ae670dd&e=edcf27e3bc>
> *Copyright ? 2018 *Marinus Analytics, All rights reserved.*
>
>
>
> Want to change how you receive these emails?
> You can update your preferences
> <https://marinusanalytics.us3.list-manage.com/profile?u=431757831d2c1c5ec10ecfdf9&id=07193289f9&e=edcf27e3bc>
> or unsubscribe from this list
> <https://marinusanalytics.us3.list-manage.com/unsubscribe?u=431757831d2c1c5ec10ecfdf9&id=07193289f9&e=edcf27e3bc&c=27d4e03631>
> .
>
> [image: Email Marketing Powered by MailChimp]
> <http://www.mailchimp.com/monkey-rewards/?utm_source=freemium_newsletter&utm_medium=email&utm_campaign=monkey_rewards&aid=431757831d2c1c5ec10ecfdf9&afl=1>
>



-- 
Andrew W. Moore <http://www.cs.cmu.edu/~awm/> | Professor and Dean, School
of Computer Science <http://www.cs.cmu.edu/> | Carnegie Mellon University
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20180607/8309b303/attachment-0001.html>

From mmv at cs.cmu.edu  Thu Jun  7 14:36:30 2018
From: mmv at cs.cmu.edu (Manuela Veloso)
Date: Thu, 7 Jun 2018 14:36:30 -0400
Subject: Sex Trafficking Victim Found Online from 2 Year Old Photo
In-Reply-To: <CAOQLAA3qaksJ7cLMW4pAUis1cpz7y5C_vTci0Ld9=Jr83hwGsg@mail.gmail.com>
References: <431757831d2c1c5ec10ecfdf9.edcf27e3bc.20180607162316.27d4e03631.f376f554@mail71.atl51.rsgsv.net>
 <CAJvAoysZrvB38B_rPZu4eKc1HNvDVKBDiJ+6VvvQHszLv=U+=g@mail.gmail.com>
 <CAOQLAA3qaksJ7cLMW4pAUis1cpz7y5C_vTci0Ld9=Jr83hwGsg@mail.gmail.com>
Message-ID: <CAF=M-rKCKjM9m3KNJZebJ+kF8soOeiPoqhWyYf_VReOE1q8OzQ@mail.gmail.com>

Ditto!!! Congratulations Artur and team! And thank you!!
Best,
        Manuela


On Thu, Jun 7, 2018 at 1:20 PM, Andrew Moore <awm at cs.cmu.edu> wrote:

> This is the perfect depiction of why I'm always so happy to get out of bed
> in the morning. Warm congratulations.
>
> Artur: this is public, so I assume it's okay to publicize this story?
>
>
> On Thu, Jun 7, 2018 at 12:50 PM, Artur Dubrawski <awd at cs.cmu.edu> wrote:
>
>> Sometimes really good things come out of our work.
>>
>> Congrats to the Traffic Jam team!
>>
>> Cheers,
>> Artur
>>
>> ---------- Forwarded message ---------
>> From: Marinus Analytics <info at marinusanalytics.com>
>> Date: Thu, Jun 7, 2018 at 12:39 PM
>> Subject: Sex Trafficking Victim Found Online from 2 Year Old Photo
>> To: <awd at cs.cmu.edu>
>>
>>
>> It started with an outcry from Allie, who was pimped 2 years ago by a
>> violent trafficker who went by the name ?Julian.?
>>
>> <https://marinusanalytics.us3.list-manage.com/track/click?u=431757831d2c1c5ec10ecfdf9&id=8cc594f33b&e=edcf27e3bc>
>> Traffic Jam Success Story
>>
>> ?Justice is something for which every generation has to strive.?
>> ? John M. Perkins
>>
>> *The Outcry*
>> It started with an outcry from Allie,* who was pimped 2 years ago by a
>> violent trafficker who went by the name ?Julian.?* She told Detective John
>> Patterson,* ?I want to get out of this, because of what Julian?s done to
>> me. And he did it to a 15 year old girl, too.? Similar cases of this
>> scope?which grew to 21 identified victims?would take a year; Detective
>> Patterson built this case in about 3 months.  Detective Patterson credits
>> the importance of good experience, trainings, and technology tools like
>> Traffic Jam.
>>
>> *The Violent Pimp*
>> Julian was a violent pimp, and he required a $1,500 per day quota for
>> each victim; he would beat them if they didn?t bring home that money each
>> night. He had been arrested many times in the past; for aggravated assault
>> with a motor vehicle, running his victims over with a car several times.
>> For strangling one of his victims until she passed out. For punching and
>> assaulting others. But force, fraud, and coercion aren?t always physical;
>> he also threatened to kill their children if they didn?t work for him.
>>
>> Julian recruited his victims in person and on social media apps like
>> Instagram. He broke down one victim by recruiting her to work as a
>> stripper, and then repeatedly raping her over a number of days. When she
>> still refused to sell sex for him, he withheld food until she broke down
>> and agreed. When she tried to escape, he used a location tracking app on
>> her phone to chase her down. He found her, assaulted her, and put her back
>> to work. She couldn?t work for a few more days, because customers would
>> notice her black eye.
>>
>> *Traffic Jam*
>> Detective Patterson used Allie?s testimony to begin piecing together the
>> case. By searching victims? Facebook photos through Traffic Jam?s
>> FaceSearch, he was able to find their ads posted across the country, from
>> the South all the way up to Pennsylvania and back. He said, ?I used Traffic
>> Jam to map out the course that Allie exactly described.?
>>
>> He was searching for Jessica,* an underage victim Allie had told him
>> about. He scrolled through Jessica?s Instagram, and found the most recent
>> pictures she posted of herself, which were over 2 years old, from when she
>> was fifteen. He said, ?I didn?t think it would lead to anything, because it
>> was such an old photo. But I thought I?d run it through FaceSearch just in
>> case. I couldn?t believe it when the 2 year old photo returned top matches
>> in FaceSearch that looked just like her.? He saw in the Traffic Jam trail
>> that she was currently posting in California, but had posted in his city in
>> the months before. Through use of FaceSearch and confirmation of the
>> locations/times in the Traffic Jam trail, Detective Patterson successfully
>> identified Jessica using a photo that was 2 years old.
>>
>> Allie told him about another victim, Sammy,* and her Facebook account. He
>> found some year-old pictures of Sammy on her Facebook profile and uploaded
>> them into FaceSearch to see if he could find a match. FaceSearch returned
>> top matches, none of which looked like Sammy. Patterson said, ?I thought
>> the matches weren?t her, they just didn?t look like her.? He sanity-checked
>> the top matches, by checking the timing and location of the ads. Then, he
>> said, ?I found that one of the phone numbers in the ad was registered to
>> her name. That made me realize that the pictures from the FaceSearch
>> results were a correct match, but I didn?t recognize her at first because
>> she had changed her appearance so drastically.? When the appearance of the
>> victim looked completely different, FaceSearch was still able to make a
>> positive match, in seconds.
>>
>> *Where Are They Now*
>> By using various technology tools like Traffic Jam, paired with victim
>> interviews and evidence gathered through search warrants, Detective
>> Patterson determined the 21 victims Julian had exploited over a number of
>> years. Patterson assembled a history of money transfers from the victims to
>> their pimp. He confirmed that many of Julian?s victims were working in
>> different states (as correlated to their ads) and wiring their earnings
>> from out of state back to Julian. He determined Julian was making about
>> $15,000 a month, from 2 girls alone; and he had a total of 21 victims over
>> the span of the investigation.
>>
>> The police department received an arrest warrant for Julian for 6
>> felonies, which they served him when he came out of the house on trash day.
>> Detectives intervened right after he had threatened to kill another
>> victim?s baby if she didn?t keep working for him. This violent trafficker
>> is currently in jail without bail, awaiting prosecution and potential life
>> in prison for his crimes, all thanks to the tireless efforts of Detective
>> Patterson and his team.
>>
>>
>> * names have been changed to protect individual?s identity
>> *Contact*
>>
>> For law enforcement who would like to be in contact with the detective
>> who submitted this story, please email us <info at marinusanalytics.com>
>> and we will connect you.
>>
>> Have a success story other investigators should know about? Share your
>> story <info at marinusanalytics.com>.
>>
>> <https://marinusanalytics.us3.list-manage.com/track/click?u=431757831d2c1c5ec10ecfdf9&id=64b84f5d80&e=edcf27e3bc>
>>
>> <https://marinusanalytics.us3.list-manage.com/track/click?u=431757831d2c1c5ec10ecfdf9&id=5c04c2ea7f&e=edcf27e3bc>
>>
>> <https://marinusanalytics.us3.list-manage.com/track/click?u=431757831d2c1c5ec10ecfdf9&id=141ae670dd&e=edcf27e3bc>
>> *Copyright ? 2018 *Marinus Analytics, All rights reserved.*
>>
>>
>>
>> Want to change how you receive these emails?
>> You can update your preferences
>> <https://marinusanalytics.us3.list-manage.com/profile?u=431757831d2c1c5ec10ecfdf9&id=07193289f9&e=edcf27e3bc>
>> or unsubscribe from this list
>> <https://marinusanalytics.us3.list-manage.com/unsubscribe?u=431757831d2c1c5ec10ecfdf9&id=07193289f9&e=edcf27e3bc&c=27d4e03631>
>> .
>>
>> [image: Email Marketing Powered by MailChimp]
>> <http://www.mailchimp.com/monkey-rewards/?utm_source=freemium_newsletter&utm_medium=email&utm_campaign=monkey_rewards&aid=431757831d2c1c5ec10ecfdf9&afl=1>
>>
>
>
>
> --
> Andrew W. Moore <http://www.cs.cmu.edu/~awm/> | Professor and Dean, School
> of Computer Science <http://www.cs.cmu.edu/> | Carnegie Mellon University
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20180607/fd5c1989/attachment-0001.html>

From awd at cs.cmu.edu  Fri Jun  8 11:16:41 2018
From: awd at cs.cmu.edu (Artur Dubrawski)
Date: Fri, 8 Jun 2018 11:16:41 -0400
Subject: Auton Lab food safety work reflected in the media
Message-ID: <CAJvAoysKgdzJzLx4oLq6O+C26vQopO+011KXgsVEx5YeocY4gA@mail.gmail.com>

https://www.delltechnologies.com/en-us/perspectives/how-robots-and-ai-are-transforming-the-food-industry/

Note that the authors erred on the count of foodborne disease cases in the
US (it is almost 70M not 7M).

Cheers
Artur
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20180608/1dce69bf/attachment.html>

From predragp at andrew.cmu.edu  Thu Jun 14 17:32:43 2018
From: predragp at andrew.cmu.edu (Predrag Punosevac)
Date: Thu, 14 Jun 2018 17:32:43 -0400
Subject: LOV5 rebooted
Message-ID: <CAEFFt-F6sFFkwvj0+6_9eN0eQzfwLYd_bavOGEd2QdTcOMxxJA@mail.gmail.com>

lov5 was killed by NFS rw activity. I had to reboot it. I apologize for the
inconvenience.

Best,
Predrag
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20180614/54469d58/attachment.html>

From awd at cs.cmu.edu  Wed Jun 20 09:30:36 2018
From: awd at cs.cmu.edu (Artur Dubrawski)
Date: Wed, 20 Jun 2018 09:30:36 -0400
Subject: Maria De Arteaga receives the Microsoft Research Dissertation Grant
Message-ID: <CAJvAoytvgTmtZXusMo+aG4hNpC-xLoYfQ_AFi81vEt=6tT356A@mail.gmail.com>

Dear Autonians,

Please join me in congratulating Maria on receiving this prestigious award!

Details can be found here:
https://www.microsoft.com/en-us/research/blog/microsoft-research-dissertation-grants-broadening-the-phd-pipeline-to-increase-innovation/

Way to go Maria!

Cheers,
Artur
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20180620/85c0ae7f/attachment.html>

From mmv at cs.cmu.edu  Wed Jun 20 14:30:15 2018
From: mmv at cs.cmu.edu (Manuela Veloso)
Date: Wed, 20 Jun 2018 14:30:15 -0400
Subject: Maria De Arteaga receives the Microsoft Research Dissertation
 Grant
In-Reply-To: <CAJvAoytvgTmtZXusMo+aG4hNpC-xLoYfQ_AFi81vEt=6tT356A@mail.gmail.com>
References: <CAJvAoytvgTmtZXusMo+aG4hNpC-xLoYfQ_AFi81vEt=6tT356A@mail.gmail.com>
Message-ID: <CAF=M-rLArFZvLCCqk35Rpw9ejY7y_Zab_SUWjMbM+87k4uWs0g@mail.gmail.com>

CONGRATULATIONS, Maria! Remarkable!! :-)
Best,
       Manuela


On Wed, Jun 20, 2018 at 9:30 AM, Artur Dubrawski <awd at cs.cmu.edu> wrote:

> Dear Autonians,
>
> Please join me in congratulating Maria on receiving this prestigious award!
>
> Details can be found here:
> https://www.microsoft.com/en-us/research/blog/microsoft-
> research-dissertation-grants-broadening-the-phd-pipeline-
> to-increase-innovation/
>
> Way to go Maria!
>
> Cheers,
> Artur
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20180620/31d3ff42/attachment.html>

From predragp at andrew.cmu.edu  Wed Jun 27 15:32:00 2018
From: predragp at andrew.cmu.edu (Predrag Punosevac)
Date: Wed, 27 Jun 2018 15:32:00 -0400
Subject: Two-Factor Authentication (2fa)
Message-ID: <20180627193200.YXMjEUfpX%predragp@andrew.cmu.edu>

Dear Autonians,

I would like to give you heads up regarding incoming changes in how you
log remotely using ssh into the Auton Lab machines.

As you know Internet is becoming one large cyber battlefield and we will
have to take things to the next level. Before the beginning of the Fall
semester I will enable DuoUnix Two-Factor Authentication (2fa) on our
gateways as well as force people to use ssh-keys instead of passwords
(unfortunately I can't enforce password protected ssh-keys which you
should do anyway).

https://duo.com/docs/duounix

I have already acquired Integration key, Secret Key, and API Hostname
from the School of Computer Science as they cost $3-$5 per user per
month depending on the service level when purchased directly from the
company. I have already tested them and it works like a charm. However
both the users and I (your system admin) will have to do few things
before I can turn on 2fa.

You are probably doing this already but if you have not done it before
you will have to register your smartphone or tablet

https://www.cmu.edu/computing/services/security/identity-access/authentication/how-to/2fa-register.html

In order to do that you will have to have valid CMU card (students,
faculty, and staff). That will create an immediate problem with the
Auton Lab affiliates which don't have CMU affiliation. I will have to
think how to resolve that problem (possibly creating less secure gateway
just for those users). I know that some of you have opted out for the
school provided devices instead of installing Dou Mobile applet. I am
not in position to accommodate such requests and in reality your smart
phone is already a spyware. If you don't install Duo Mobile applet you
will have physically come to school and use one of our desktops
(although I am not sure how long you will have to be able to use them
without CAC).    

On my side I have to make sure that all usernames and e-mail addresses
in the Auton Lab LDAP database are exactly the same as your CMU Andrew
userid. I know at least one Auton Lab account holder (me) whose Andrew
userid was different (until today) than his Auton Lab userid.

Thank you for your kind cooperation and patience with this matter.

Sincerely,
Predrag Punosevac