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 2027: March 23–26, 2027
We are pleased to announce the Fourth Conference on Parsimony and Learning, to be held at Hitotsubashi Hall in Tokyo, Japan.
Call for Papers
CPAL 2027 invites theoretical, methodological, empirical, and systems contributions that make parsimony central to learning—through sparsity, low rank, symmetry, modularity, compressibility, or structured computation. We welcome work on foundation, generative, multimodal, and agentic models when the underlying parsimony principle is explicit. Submit original, unpublished research to the Proceedings Track, with accepted papers published in PMLR, or timely work at different stages of development to the Recent Spotlight Track, presented without archival proceedings. We value technically sound research with clear insights for the community; acceptance does not require a new benchmark state of the art or novelty for its own sake.
Read the full Call for Papers · Submission Tracks · Key Dates
Important Dates
- Abstract registration November 23, 2026
- Proceedings paper deadline December 5, 2026
- Recent Spotlight submission deadline January 18, 2027
All three deadlines are 23:59 Anywhere on Earth (AoE). View the full schedule for review, notification, and other conference dates.
Conference Leadership
General Chairs
-
Yi Ma
The University of Hong Kong
-
Taiji Suzuki
The University of Tokyo & RIKEN AIP
Conference Sponsors
CPAL 2027 welcomes support from industry, foundations, research organizations, and other partners. See the Sponsors page and the CPAL 2027 Sponsorship Prospectus for sponsorship opportunities.