Kernel Alignment for Unsupervised Feature Selection via Matrix Factorization

no code implementations • 13 Mar 2024 • Ziyuan Lin, Deanna Needell

By doing so, our model can learn both linear and nonlinear similarity information and automatically generate the most appropriate kernel.

feature selection

Comparison of Deep Learning Segmentation and Multigrader-annotated Mandibular Canals of Multicenter CBCT scans

no code implementations • 27 May 2022 • Jorma Järnstedt, Jaakko Sahlsten, Joel Jaskari, Kimmo Kaski, Helena Mehtonen, Ziyuan Lin, Ari Hietanen, Osku Sundqvist, Vesa Varjonen, Vesa Mattila, Sangsom Prapayasotok, Sakarat Nalampang

Deep learning approach has been demonstrated to automatically segment the bilateral mandibular canals from CBCT scans, yet systematic studies of its clinical and technical validation are scarce.

Out-of-Distribution Generalization

Peacock Bundles: Bundle Coloring for Graphs with Globality-Locality Trade-off

no code implementations • 2 Sep 2016 • Jaakko Peltonen, Ziyuan Lin

In large graphs the necessary compromise is whether to differentiate colors sharply between locally occurring strongly bundled edges ("local bundles"), or also between the weakly bundled edges occurring globally over the graph ("global bundles"); we allow a user-set global-local tradeoff.

Dimensionality Reduction

An Information Retrieval Approach to Finding Dependent Subspaces of Multiple Views

no code implementations • 19 Nov 2015 • Ziyuan Lin, Jaakko Peltonen

The basic CCA is restricted to maximizing a simple dependency criterion, correlation, measured directly between data coordinates.

Information Retrieval Retrieval