Third Workshop on Seeking Low‑Dimensionality in Deep Neural Networks

January 2023, MBZUAI

Join Us in Abu Dhabi from 3-7 January 2023!

Video Recordings from the Workshop

Workshop Schedule

Announcing the workshop venue: W Abu Dhabi Hotel at Yas Island (Google Maps)

Register here for in-person and remote participation

Key Dates and Deadlines

• Workshop dates: 3rd - 7th January, 2023; in person at W Hotel, Abu Dhabi
• Submission portal opens: October 1st, 2022
• Paper submission deadline: November 6th, 2022
• Acceptance and travel grant notification: November 20th, 2022

The top accepted papers will be recommended for inclusion in a future special issue of the IEEE Journal of Selected Topics in Signal Processing.

Workshop Themes

In the past decade, deep learning has demonstrated unprecedented performance across many different domains. Modern neural networks learn high-performing nonlinear representations for complex multi-modal data (e.g. text and image) without labels; they generate near-photorealistic high-resolution images from random noise; and they efficiently reconstruct data from compressive, corrupted measurements in various imaging and structural prediction tasks.

However, despite recent endeavors, the underlying principles behind their success remain shrouded in mystery. At a fundamental level, each of these successes, and numerous others in scientific and engineering applications, stem from the low-dimensional structure present both in the training data and in the networks themselves. This presents a significant opportunity to bring insights from better-understood low-dimensional models to the setting of deep learning, where models are notoriously plagued by issues of poor resource and data efficiency, robustness, and generalization.

Given these exciting but relatively less-exploited connections, this workshop aims to bring together experts in machine learning, applied mathematics, signal processing, and optimization, to share recent progress and foster collaborations on the mathematical foundations of deep learning. As the community continues to embrace the power of deep learning, with unprecedented new challenges in terms of modeling and interpretability, we hope to stimulate vibrant discussions towards bridging the gap between the theory and practice of deep learning, with low-dimensionality as a unifying focus.

Confirmed Speakers

Information on the speakers’ planned talks is available here.

Misha Belkin

UC San Diego

Beidi Chen

Meta

Ivan Dokmanić

University of Basel

Simon Du

University of Washington

Reinhard Heckel

TU Munich

Gitta Kutyniok

LMU Munich

Jason Lee

Princeton

Qi Lei

NYU

Yi Ma

UC Berkeley

Rina Panigrahy

Google

Tomaso Poggio

MIT

Qing Qu

UMich

Ohad Shamir

Weizmann Institute

Mahdi Soltanolkotabi

USC

Daniel Soudry

Technion

Weijie Su

UPenn

René Vidal

Johns Hopkins

Yu-Xiang Wang

UCSB

Bihan Wen

NTU

Fanny Yang

ETH Zurich

Workshop Schedule

All times in GST (GMT+4).

Morning sessions 9 AM - 12 PM; Afternoon sessions 1:30 PM - 6 PM

Organizers

Yi Ma

UC Berkeley

General Chair

Eric P. Xing

MBZUAI

General Chair (Local)

Samuel Horváth

MBZUAI

Local Chair

Karthik Nandakumar

MBZUAI

Local Chair

Martin Takáč

MBZUAI

Local Chair

Qing Qu

University of Michigan

Program Chair and Publication Chair

Jeremias Sulam

Johns Hopkins University

Program Chair

Atlas Wang

UT Austin

Program Chair

John Wright

Columbia University

Program Chair

Yuqian Zhang

Rutgers University

Program Chair

Yuejie Chi

Carnegie Mellon University

Publicity Chair

Chong You

Google NYC

Publication Chair

Zhihui Zhu

Ohio State University

Publication Chair

Sam Buchanan

TTIC

Tutorial Chair

Saiprasad Ravishankar

Michigan State University

Online Chair

Host Institution