[Colloquium] Owen Melia MS Presentation/Mar 3, 2023

[Megan Woodward]

This is an announcement of Owen Melia's MS Presentation
===============================================
Candidate: Owen Melia

Date: Friday, March 03, 2023

Time:  3:45 pm CST

Location: JCL 223

M.S. Paper Title: ROTATION-INVARIANT RANDOM FEATURES PROVIDE A STRONG BASELINE FOR MACHINE LEARNING ON 3D POINT CLOUDS

Abstract: Rotational invariance is a popular inductive bias used by many fields in machine learning,
such as computer vision and machine learning for quantum chemistry. Rotation-invariant
machine learning methods set the state of the art for many tasks, including molecular property
prediction and 3D shape classification. These methods generally either rely on taskspecific
rotation-invariant features, or they use general-purpose deep neural networks which
are complicated to design and train. We suggest a simple and general-purpose method for
learning rotation-invariant functions of three-dimensional point cloud data using a random
features approach. Specifically, we extend the random features method of Rahimi and Recht
[2007] by deriving a version that is invariant to three-dimensional rotations and showing
that it is fast to evaluate on point cloud data. We show through experiments that our
method matches or outperforms the performance of general-purpose invariant architectures
on standard molecular property prediction benchmark datasets QM7 and QM9. We also
show that our method is competitive with other general-purpose invariant architectures on
the ModelNet40 shape classification task. Finally, we show that our method is an order of
magnitude faster at prediction time than competing kernel methods.

Advisors: Rebecca Willett

Committee Members: Risi Kondor, Eric Jonas, and Rebecca Willett




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.cs.uchicago.edu/pipermail/colloquium/attachments/20230302/76db848d/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Owen_MS_thesis.pdf
Type: application/pdf
Size: 1291071 bytes
Desc: Owen_MS_thesis.pdf
URL: <http://mailman.cs.uchicago.edu/pipermail/colloquium/attachments/20230302/76db848d/attachment-0001.pdf>