[Colloquium] [CDAC] February 1 - Mukund Sundararajan (Google)

[Rob Mitchum]

*CDAC Distinguished Speaker Series*


*Mukund Sundararajan*

*Principal Research Scientist/Director*
*Google*

*Monday, February 1st*
*3:00 p.m. - 4:00 p.m.*
*Zoom (RSVP for login
<https://www.eventbrite.com/e/cdac-distinguished-speaker-series-mukund-sundararajan-google-tickets-129939760321>)
or YouTube <https://youtu.be/rJ1agP_mLRQ> (no registration required)*


*Using Attribution to Understand Deep Neural Networks*
There was a neural model for predicting cancer from XRays. It had good
accuracy on held out training data. But when we attributed its predictions
back to the pixels of the XRays, we found that the network relied on barely
visible pen marks that the doctors had made on the training data, and not
the pathology of cancer. Naturally, the model was not deployed!

I work on techniques to perform prediction attribution of this kind. The
target of the attribution can be input features (pixels in the example
above), or interactions between its input features, or neurons. or training
data examples. Attributions are reductive; i.e, they abstract away most of
the interactions and a lot of the non-linearity of neural networks.
However, attributions, done systematically, are effective at uncovering
bugs as in the anecdote above.

We will briefly discuss the theory (e.g connections to the Taylor series,
Shapley values, and Stochastic Gradient Descent) and philosophy of
attribution, and other amusing examples of bugs.

If you are a deep learning practitioner, you can easily apply attribution
to your own models; all the techniques can be implemented with less than
ten lines of code.

*Bio: *I am a principal research scientist/director at Google. These days,
I analyze complex machine learning models. I have also worked on
question-answering systems, ad auctions, security protocol analysis,
privacy, and computational biology.

There once was a RS
<https://urldefense.com/v3/__https://acronyms.thefreedictionary.com/Research*scientist__;Kw!!BpyFHLRN4TMTrA!rFEPEjUNwjV1j2QJX8KpDgAr9IP7X2EgHMFaplOMI4daPSiVBCOnqpFRzmtRM-OETH7WEA$>
called MS
<https://urldefense.com/v3/__https://sites.google.com/site/sundararajanmukund/__;!!BpyFHLRN4TMTrA!rFEPEjUNwjV1j2QJX8KpDgAr9IP7X2EgHMFaplOMI4daPSiVBCOnqpFRzmtRM-Ng_ZkGfw$>
,
He studies models
<https://urldefense.com/v3/__https://en.wikipedia.org/wiki/Machine_learning__;!!BpyFHLRN4TMTrA!rFEPEjUNwjV1j2QJX8KpDgAr9IP7X2EgHMFaplOMI4daPSiVBCOnqpFRzmtRM-MCQ6d1rw$>
that are a mess,
A director at Google.
Accurate and frugal,
Explanations are what he likes best.


*Part of the CDAC Winter 2021 Distinguished Speaker Series
<https://cdac.uchicago.edu/news/announcing-the-cdac-winter-2021-distinguished-speaker-series/>:*

*Bias Correction: Solutions for Socially Responsible Data Science*
Security, privacy and bias in the context of machine learning are often
treated as binary issues, where an algorithm is either biased or fair,
ethical or unjust. In reality, there is a tradeoff between using technology
and opening up new privacy and security risks. Researchers are developing
innovative tools that navigate these tradeoffs by applying advances in
machine learning to societal issues without exacerbating bias or
endangering privacy and security. The CDAC Winter 2021 Distinguished
Speaker Series will host interdisciplinary researchers and thinkers
exploring methods and applications that protect user privacy, prevent
malicious use, and avoid deepening societal inequities — while diving into
the human values and decisions that underpin these approaches.

-- 
*Rob Mitchum*

*Associate Director of Communications for Data Science and Computing*
*University of Chicago*
*rmitchum at uchicago.edu <rmitchum at ci.uchicago.edu>*
*773-484-9890*
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.cs.uchicago.edu/pipermail/colloquium/attachments/20210126/2f74b1f9/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: mukund-slide1.png
Type: image/png
Size: 442167 bytes
Desc: not available
URL: <http://mailman.cs.uchicago.edu/pipermail/colloquium/attachments/20210126/2f74b1f9/attachment-0001.png>