Fwd: PhD Speaking Qualifier: Design with Interpretability in Mind: An Alternate Ethos for Data Science

[Artur Dubrawski]

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

Barnabás Póczos

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>