Finally, we propose a Channel-Temporal Attention Network (CTAN) to integrate these blocks into existing architectures.
Under the team name xy9, our submission achieved 5th place in terms of top-1 accuracy for verb class and all top-5 accuracies.
Machine learning is a general-purpose technology holding promises for many interdisciplinary research problems.
Because H2NT can sparsify networks with motif structures, it can also improve the computational efficiency of existing network embedding methods when integrated.
Heterogeneous graph representation learning aims to learn low-dimensional vector representations of different types of entities and relations to empower downstream tasks.
no code implementations • 5 Jun 2020 • Markus D. Schirmer, Archana Venkataraman, Islem Rekik, Minjeong Kim, Stewart H. Mostofsky, Mary Beth Nebel, Keri Rosch, Karen Seymour, Deana Crocetti, Hassna Irzan, Michael Hütel, Sebastien Ourselin, Neil Marlow, Andrew Melbourne, Egor Levchenko, Shuo Zhou, Mwiza Kunda, Haiping Lu, Nicha C. Dvornek, Juntang Zhuang, Gideon Pinto, Sandip Samal, Jennings Zhang, Jorge L. Bernal-Rusiel, Rudolph Pienaar, Ai Wern Chung
A second set of 100 subjects (50 neurotypical controls, 25 ADHD, and 25 ASD with ADHD comorbidity) was used for testing.
The use of drug combinations often leads to polypharmacy side effects (POSE).
Ranked #1 on Pose Prediction on SUN-Mem
We use public data to construct 13 transfer learning tasks in brain decoding, including three interesting multi-source transfer tasks.
This paper proposes a new Mixed-Order Spectral Clustering (MOSC) approach to model both second-order and third-order structures simultaneously, with two MOSC methods developed based on Graph Laplacian (GL) and Random Walks (RW).
While functional magnetic resonance imaging (fMRI) is important for healthcare/neuroscience applications, it is challenging to classify or interpret due to its multi-dimensional structure, high dimensionality, and small number of samples available.