Blar i NTNU Open på forfatter "Yu, Tong"
-
Classification of long sequential data using circular dilated convolutional neural networks
Cheng, Lei; Khalitov, Ruslan; Yu, Tong; Zhang, Jing; Yang, Zhirong (Peer reviewed; Journal article, 2023)Classification of long sequential data is an important Machine Learning task and appears in many application scenarios. Recurrent Neural Networks, Transformers, and Convolutional Neural Networks are three major techniques ... -
Paramixer: Parameterizing Mixing Links in Sparse Factors Works Better than Dot-Product Self-Attention
Yu, Tong; Khalitov, Ruslan; Cheng, Lei; Yang, Zhirong (Journal article, 2022)Self-Attention is a widely used building block in neural modeling to mix long-range data elements. Most self-attention neural networks employ pairwise dot-products to specify the attention coefficients. However, these ... -
Sparse factorization of square matrices with application to neural attention modeling
Khalitov, Ruslan; Yu, Tong; Cheng, Lei; Yang, Zhirong (Peer reviewed; Journal article, 2022)Square matrices appear in many machine learning problems and models. Optimization over a large square matrix is expensive in memory and in time. Therefore an economic approximation is needed. Conventional approximation ... -
Stochastic Local Search for Efficient Hybrid Feature Selection
Mengshoel, Ole Jakob; Yu, Tong; Riege, Jon; Flogard, Eirik Lund (Chapter, 2021)There is a need to study not only accuracy but also computational cost in machine learning. Focusing on both accuracy and computational cost of feature selection, we develop and test stochastic local search (SLS) heuristics ... -
Stochastic Local Search Heuristics for Efficient Feature Selection: An Experimental Study
Mengshoel, Ole Jakob; Flogard, Eirik Lund; Riege, Jon; Yu, Tong (Peer reviewed; Journal article, 2021) -
Understanding and Improving Recurrent Networks for Human Activity Recognition by Continuous Attention
Zeng, Ming; Gao, Haoxiang; Yu, Tong; Mengshoel, Ole Jakob; Langseth, Helge; Lane, Ian; Liu, Xiaobing (Chapter, 2018)Deep neural networks, including recurrent networks, have been successfully applied to human activity recognition. Unfortunately, the final representation learned by recurrent networks might encode some noise (irrelevant ... -
Understanding the Cost of Fitness Evaluation for Subset Selection: Markov Chain Analysis of Stochastic Local Search
Mengshoel, Ole Jakob; Flogard, Eirik Lund; Yu, Tong; Riege, Jon (Chapter, 2022)With a focus on both the fitness and cost of subset selection, we study stochastic local search (SLS) heuristics in this paper. In particular, we consider subset selection problems where the cost of fitness function ...