Call for Papers
We welcome all original research papers of up to 4 pages in length,
using the template provided below. This length does not include
references or any supplementary materials. Reviewers are not obliged
to read supplementary materials when reviewing the paper. Submissions
should be a single file in .pdf format. The review process is
double-blind, so please ensure that all papers are appropriately
anonymized.
This workshop is non-archival; even though all accepted papers will be available on OpenReview, there are no formally-published proceedings.
| Submission link | OpenReview |
| Submission deadline |
Scope and topics
Please find a list of topics of interest, sorted alphabetically. If you are not sure whether your topic might be a good fit for the workshop, feel free to contact us at geometry@topology.rocks.
-
Approximations: Can we approximate parts of the computational pipeline in topological data analysis, for example when calculating Vietoris–Rips complexes and their associated filtrations, or when calculating distances between persistence diagrams?
-
Benchmark data sets and software: What are suitable benchmark data sets to quickly compare different topological data analysis, graph representation learning, and manifold learning methods in a reproducible manner? What are “our” MNIST or CIFAR data sets, and how do we go beyond them? How do the existing software tools in these fields measure up for ever-increasing data, and how do we address the software needs?
-
Scaling and parallel processing: How can we scale implementations to the dataset sizes that we encounter in machine learning, with bigger graphs, data manifolds, or collections of many geometric “objects”?
-
Current and future applications: In which projects and applications does the incorporation of topology and geometry information have decisive edge over alternative approaches, and how to identify relevant high-impact “niches” methodically?
-
Feature descriptors: How can we employ geometric and topological descriptors in unsupervised or supervised machine learning scenarios? Which structural patterns are captured, or missed, by them and how can we improve their integration into representation learning frameworks?
-
Higher-order features: In many applications, the theoretical richness of extracted geometric and topological features is limited in practice (e.g., restricting information to connected components, local similarity, and smoothness). What can we say about higher-order regularities or patterns? What would be the structural equivalent of considering textures rather than edges in images? When are they suitable, and how can we approximate these efficiently?
Style
We only accept submissions that have been prepared using LaTeX. Use the following workshop style files for your submission: