Teacher Assistant-Based Knowledge Distillation Extracting Multi-level Features on Single Channel Sleep EEG

Authors: Heng Liang, Yucheng Liu, Haichao Wang, Ziyu Jia

IJCAI 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We evaluate our method on two public sleep datasets (Sleep-EDF and ISRUC-III). Compared to the baseline methods, the results show that our knowledge distillation framework achieves state-of-the-art performance. Experimental results show that our knowledge distillation framework achieves SOTA performance compared to existing knowledge distillation methods.
Researcher Affiliation Academia 1 Brainnetome Center, Institute of Automation, Chinese Academy of Sciences, Beijing, China 2 Tsinghua-Berkeley Shenzhen Institute, Shenzhen, China
Pseudocode No The paper does not contain any structured pseudocode or algorithm blocks.
Open Source Code Yes The source code is available at https:// github.com/Hychao Wang/Sleep KD.
Open Datasets Yes We evaluate our method on two public datasets: ISRUC-III [Khalighi et al., 2016] and Sleep-EDF [Kemp et al., 2018].
Dataset Splits Yes We split the datasets into the train, validation, and test sets by a ratio of 8:1:1 on Sleep-EDF and ISRUC-III separately.
Hardware Specification No The paper does not provide specific hardware details such as GPU models, CPU types, or memory specifications used for running the experiments. It only mentions implementing models with 'Tensor Flow'.
Software Dependencies No The paper mentions implementing models with 'Tensor Flow' but does not specify a version number or any other software dependencies with their versions, which is required for reproducible description.
Experiment Setup Yes We use Adam as the optimizer in each experiment. In experiments of the CNN framework, we choose Salient Sleep Net as a representative. The learning rate of Salient Sleep Net is 0.001. The number of training epochs is 60 and the batch size is 8. The weights are α = 0.3, β = 0.2, γ = 0.4 and δ = 0.1. In experiments of the CNN and RNN framework, we choose Deep Sleep Net as a representative. Deep Sleep Net has a learning rate of 0.00001. The number of training epochs is 200 for Sleep EDF, 300 for ISRUC-III and the batch size is 20. The weights are α = 1.0, β = 0.1 and γ = δ = 1.0.