Excited to share my 1st paper as a PhD student at @UWMadPhysics working with @datascience_uw!
Our paper introduces a novel way to benchmark self-supervised ML models and evaluate mutual information estimators on realistic data.
Glad to have worked with this group. More to come!
New paper, with @RaheemHashmani@merz_garrett@marielpettee@_helenqu
We introduce a framework for generating realistic, highly multimodal datasets with explicitly calculable mutual information. This is helpful for studying self-supervised learning
https://t.co/XliwutGTNi
From a friend of mine
"I opted out of @Dominos' selling of my info. Not only did it take 4 different screens, 1 email, 1 SMS, and 6 business days, they got back to me with this message."
How can that even be legal. We will continue to sell your data if you clear your cookies?
Yesterday, we also had the
@uwmadphy Physics
Fair. Great job Sarah organizing it all!
Here we can see @merz_garrett, who works with me at @datascience_uw, manning the Cloud Chamber exhibit!
Was able to get a photo with @LibnOfCongress. Dr. Carla Hayden is the 14th Librarian of Congress and the first woman and African American to hold the position! She’s a trailblazer in making knowledge and history accessible to everyone. @librarycongress
The second one is the Cathode-Ray Tube (CRT). A vacuum tube where electrons are accelerated and deflected to create images on a fluorescent screen. Once central to TVs and monitors, it was originally key to J. J. Thomson’s experiments that helped discover the electron!
On to the next event for today!
As a @UWMadPhysics Outreach Fellow, I’m at @MadisonEastHS performing a couple modern physics experiments with other Fellows.
The first one is the Photoelectric Effect! Einstein proposed that photons come in packets with a specific energy. If the energy is large enough, it can knock an electron loose. This discovery helped him win a Nobel prize and is now easily replicated across the globe.
Excited to host the @uwmadison@NSF HDR ML Challenge kickoff meeting today with @uw_wipac. The challenge: develop algorithms that can identify subtle differences and anomalies across three provided datasets. Thanks to everyone who participated! @a3d3institute
There are a few seats left in the NSF HDR ML Challenge kickoff meeting. Participants will use ML to discover anomalies in data and compete for a cash prize. Open to
@UWMadison students, faculty, & staff, and community members. Learn more and register: https://t.co/bj76R3j5xc
I have this book! I can see the Washburn Observatory from my office. And last week @RaheemHashmani organized for the Data Science Institute to visit the old planetarium, which got me thinking about how the notion of a physics-inspired or data-driven “model” has evolved
Congrats to PhD student Raheem Hashmani on the publication of his first-author article, "A comparison of deep learning models for proton background rejection with the AMS electromagnetic calorimeter," in @MLSTjournal! The work was completed at @METU_ODTU
⬇️https://t.co/x9cmrmmcD3