Fwd: RI Ph.D. Thesis Defense: Mononito Goswami

[Artur Dubrawski]

Please mark your calendars and join Mononito at this big event. It promises
to be an intellectual treat.

Cheers,
Artur

---------- Forwarded message ---------
From: Suzanne Muth <lyonsmuth at cmu.edu>
Date: Mon, Apr 7, 2025, 4:39 PM
Subject: RI Ph.D. Thesis Defense: Mononito Goswami
To: RI People <ri-people at andrew.cmu.edu>


*RI Events Calendar Posting
<https://www.ri.cmu.edu/event/towards-pragmatic-time-series-intelligence-2/>*

*Date:* Thursday, 17 April 2025
*Time:* 4:30 p.m. (ET)
*Location:* CIC LL06 (see attached figure)
*Zoom Link:*
https://cmu.zoom.us/j/91581272900?pwd=DCOr0EFMMTuAncqaeJvNJgw9W8Xsqb.1&jst=2
*Type:* Ph.D. Thesis Defense
*Who:* Mononito Goswami
*Title:* Towards Pragmatic Time Series Intelligence

*Abstract:*
This thesis aims to democratize time series intelligence by making advanced
modeling capabilities accessible to users without specialized machine
learning knowledge. We pursue this goal through three complementary
contributions that build foundation models, improve our understanding of
them, and address challenges emerging in their practical use.

We start by introducing MOMENT, the first family of open source time series
foundation models capable of performing well on a variety of tasks on data
from diverse domains with minimal supervision. We extend these models to
handle long multivariate contexts and integrate multimodal data, enabling
their application to complex real-world scenarios where traditional
approaches fall short.

Next, we examine what these foundation models learn by investigating their
compositional reasoning abilities, representation structures, and encoded
concepts. We identify practical insights that improve both our
understanding of the models and their performance.

Then, we tackle deployment challenges by developing methods to learn from
distributed unlabeled data, assess label quality, and select robust models
when labeled data is scarce. Lastly, we explore how Large Language Model
agents can automate the time series intelligence engineering process, using
open-source tools and tools developed in this thesis.

We demonstrate the utility of our methodology in clinical settings, where
time series data is plentiful and where modeling of it can be impactful. We
conclude that specialized foundation models, combined with practical tools
supporting their real-world deployment, can substantially advance time
series intelligence and yield practical solutions of societal importance.

*Thesis Committee Members:*
Artur Dubrawski (Chair)
Jean Oh
Barnabás Póczos
Frederic Sala (University of Wisconsin-Madison)
Laurent Callot (Amazon)

Draft of the Thesis Document
<https://drive.google.com/drive/folders/1yzdjxg0Koj92msch3kRxJDemi29E9Snj?usp=sharing>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20250407/17593f5e/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: CIC Location_LL06 Lower Level.png
Type: image/png
Size: 271016 bytes
Desc: not available
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20250407/17593f5e/attachment-0001.png>