Conference on Parsimony and Learning (CPAL)

January 2024, HKU

The Conference on Parsimony and Learning (CPAL) is an annual research conference focused on addressing the parsimonious, low dimensional structures that prevail in machine learning, signal processing, optimization, and beyond. We are interested in theories, algorithms, applications, hardware and systems, as well as scientific foundations for learning with parsimony.

CPAL 2024 will take place from January 3–6
at the University of Hong Kong!

Register to Attend CPAL 2024

All CPAL attendees are required to register. The deadline to register is December 15th, 2023.

Register Now

Keynote Speakers

Information on the speakers’ planned talks is available here.

Dan Alistarh

Institute of Science and Technology Austria / Neural Magic

SueYeon Chung

New York University / Flatiron Institute

Kostas Daniilidis

University of Pennsylvania

Maryam Fazel

University of Washington

Tom Goldstein

University of Maryland

Yingbin Liang

Ohio State University

Dimitris Papailiopoulos

University of Wisconsin-Madison

Stefano Soatto

University of California, Los Angeles

Jong Chul Ye

Korea Advanced Institute of Science and Technology (KAIST)

Tentative Program

All times below are in HKT (GMT+8).

• 08:00 AM
• 08:30 AM
• 09:00 AM
• 09:30 AM
• 10:00 AM
• 10:30 AM
• 11:00 AM
• 11:30 AM
• 12:00 PM
• 12:30 PM
• 01:00 PM
• 01:30 PM
• 02:00 PM
• 02:30 PM
• 03:00 PM
• 03:30 PM
• 04:00 PM
• 04:30 PM
• 05:00 PM
• 05:30 PM
• 06:00 PM
• 06:30 PM
• 07:00 PM
• 07:30 PM
• 08:00 PM
• 08:30 PM
• 09:00 PM

• Day 1 (Jan 3)
  Wednesday

  • Registration
    8:00 AM–9:15 AM
    TBA
  • Opening Remarks
    9:15 AM–10:00 AM
    TBA
  • Stefano Soatto
    10:00 AM–11:00 AM
    TBA
    "Representation and Control of Meanings in Large Language Models and Multimodal Foundation Models"
  • Coffee Break
  • Rising Stars Session 1
    11:20 AM–12:30 PM
    TBA
  • Lunch Break
  • Maryam Fazel
    1:30 PM–2:30 PM
    TBA
    "Flat Minima and Generalization in Learning: The Case of Low-rank Matrix Recovery"
  • Oral Session 1
    2:30 PM–3:40 PM
    TBA
  • Coffee Break
  • Dan Alistarh
    4:00 PM–5:00 PM
    TBA
    "Accurate Model Compression at GPT Scale"
  • Reception
    5:00 PM–6:30 PM
    TBA

• Day 2 (Jan 4)
  Thursday

  • Yoga Meditation
    8:30 AM–9:00 AM
    TBA
  • Jong Chul Ye
    9:00 AM–10:00 AM
    TBA
    "Scaling Diffusion Model for Inverse Problems Beyond 2D and Known Forward Models"
  • Rising Stars Session 2
    10:00 AM–11:00 AM
    TBA
  • Coffee Break
  • Oral Session 2
    11:20 AM–12:30 PM
    TBA
  • Lunch Break
  • Kostas Daniilidis
    1:30 PM–2:30 PM
    TBA
    "Parsimony through Equivariance"
  • Oral Session 3
    2:30 PM–4:00 PM
    TBA
  • Coffee Break
  • Panel Discussion
    4:10 PM–5:00 PM
    TBA
  • Spotlight Poster Session 1
    5:00 PM–6:30 PM
    TBA
  • Banquet
    7:00 PM–9:00 PM
    TBA

• Day 3 (Jan 5)
  Friday

  • Yoga Meditation
    8:30 AM–9:00 AM
    TBA
  • Yingbin Liang
    9:00 AM–10:00 AM
    TBA
    "In-Context Convergence of Transformers"
  • Rising Stars Session 3
    10:00 AM–11:00 AM
    TBA
  • Coffee Break
  • Oral Session 4
    11:20 AM–12:30 PM
    TBA
  • Lunch Break
  • Dimitris Papailiopoulos
    1:30 PM–2:30 PM
    TBA
    "Teaching arithmetic to small language models"
  • Oral Session 5
    2:30 PM–3:40 PM
    TBA
  • Coffee Break
  • Oral Session 6
    4:00 PM–5:00 PM
    TBA
  • Spotlight Poster Session 2
    5:00 PM–6:30 PM
    TBA
  • Tram Tour (Tentative)
    7:00 PM–9:00 PM
    TBA

• Day 4 (Jan 6)
  Saturday

  • Yoga Meditation
    8:30 AM–9:00 AM
    TBA
  • Tom Goldstein
    9:00 AM–10:00 AM
    TBA
    "Statistical methods for addressing safety and security issues of generative models"
  • SueYeon Chung
    10:00 AM–11:00 AM
    TBA
    "Multi-level theory of neural representations:
Capacity of neural manifolds in biological and artificial neural networks"
  • Coffee Break
  • Open-Door Roundtable (Organization Committee)
    11:20 AM–12:30 PM
    TBA
  • Lunch Break
  • OPEN TO PUBLIC: Half-Day Tutorials, Part I
    (Two Parallel Tracks: Low-Dimensional Structure in Deep Networks; Inverse Problems)
    1:30 PM–3:40 PM
    TBA
  • Coffee Break
  • OPEN TO PUBLIC: Half-Day Tutorials, Part II
    (Two Parallel Tracks: Low-Dimensional Structure in Deep Networks; Inverse Problems)
    4:00 PM–6:30 PM
    TBA

• Day 5 (Jan 7)
  Sunday

  • Social Event / Tour / Etc. (TBA)
    10:00 AM–2:00 PM
    TBA

Conference Sponsors

Local Host and Main Sponsor

Platinum Sponsor

Gold Sponsors

Silver Sponsors

Rising Star Award Sponsor

Yoga Sponsor