2026-09-19T07:49:31Z https://www.ideals.illinois.edu/oai-pmh

oai:www.ideals.illinois.edu:2142/124593 2026-01-14 col_2142_5131 col_2142_8888 com_2142_5130 com_2142_8887 com_2142_234

Hasegawa-Johnson, Mark 2024-05 application/pdf en text Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2026-05-01 The student, Kai Chieh Chang, accepted the attached license on 2024-04-29 at 13:15. The student, Kai Chieh Chang, submitted this Thesis for approval on 2024-04-29 at 13:22. This Thesis was approved for publication on 2024-04-30 at 15:23. DSpace SAF Submission Ingestion Package generated from Vireo submission #20692 on 2024-09-16 at 00:44:51 Fusing multimodal neural networks: a study on sleep classification and sound event localization and detection Chang, Kai Chieh 2024-04-30 Multimodal Machine Learning This is an explorative study of multimodal large-scale transformer networks. The thesis explores methods to pretrain and fuse multiple large-scale transformer networks each responsible for a modality, in order to improve their performance on different tasks. Specifically, we first dive into the task of infant sleep classification using audio, electrocardiogram (ECG), and inertial measurement unit (IMU). We explore various pretraining and finetuning schemes, as well as different fusion techniques. We also assess the effectiveness of fusion by cross-attention with sound event localization and detection (SELD), a multichannel machine learning task with multiple outputs. We show that this multimodal network structure is generic enough to work in various settings. Text eng https://hdl.handle.net/2142/124593 Copyright 2024 Kai Chieh Chang M.S. Thesis Electrical & Computer Engr University of Illinois at Urbana-Champaign Electrical & Computer Eng