S
Seungjin Choi
Researcher at Pohang University of Science and Technology
Publications - 303
Citations - 8294
Seungjin Choi is an academic researcher from Pohang University of Science and Technology. The author has contributed to research in topics: Non-negative matrix factorization & Blind signal separation. The author has an hindex of 47, co-authored 303 publications receiving 7181 citations. Previous affiliations of Seungjin Choi include South China University of Technology & University of Notre Dame.
Papers
More filters
Posted Content
Set Transformer: A Framework for Attention-based Permutation-Invariant Neural Networks
TL;DR: This work presents an attention-based neural network module, the Set Transformer, specifically designed to model interactions among elements in the input set, and reduces the computation time of self-attention from quadratic to linear in the number of Elements in the set.
Proceedings Article
Gradient-Based Meta-Learning with Learned Layerwise Metric and Subspace
Yoonho Lee,Seungjin Choi +1 more
TL;DR: In this article, a task-specific learner of an EMMT-net performs gradient descent with respect to a meta-learned distance metric, which warps the activation space to be more sensitive to task identity.
Proceedings Article
Set Transformer: A Framework for Attention-based Permutation-Invariant Neural Networks
TL;DR: The Set Transformer as discussed by the authors is an attention-based neural network module, specifically designed to model interactions among elements in the input set, consisting of an encoder and a decoder, both of which rely on attention mechanisms.
Proceedings ArticleDOI
Convolutional neural networks for human activity recognition using multiple accelerometer and gyroscope sensors
Sojeong Ha,Seungjin Choi +1 more
TL;DR: This paper employs both partial weight sharing and full weight sharing for the CNN models in such a way that modality-specific characteristics as well as common characteristics across modalities are learned from multi-modal (or multi-sensor) data and are eventually aggregated in upper layers.
Journal ArticleDOI
Composite Common Spatial Pattern for Subject-to-Subject Transfer
TL;DR: Modifications of CSP for subject-to-subject transfer are presented, where a linear combination of covariance matrices of subjects in consideration are exploited, leading to composite CSP.