Name
Title | ||
|---|---|---|
Orthogonal features-based EEG signal denoising using fractionally compressed autoencoder |
Abstract | ||
|---|---|---|
AbstractHighlights •We introduce a fractionally based compressed auto-encoder to address the problem of EEG Signal denoising.•Tchebichef moments of the original signal have been used as feature input to the neural network, which results in conversion to sparse signals helping in accelerating the training of the auto-encoder.•Randomized singular valued decomposition is used for compressing the auto-encoder by finding the low-rank approximation of trainable weights. The compression rate is varied during evaluation to note the performance of the architecture.•The fractional order along with compression performs with similar effectiveness as a traditional auto-encoder without compression. AbstractA fractional-based compressed auto-encoder architecture has been introduced to solve the problem of denoising electroencephalogram (EEG) signals. The architecture makes use of fractional calculus to calculate the gradients during the back-propagation process, as a result of which a new hyper-parameter in the form of fractional order α has been introduced which can be tuned to get the best denoising performance. Additionally, to avoid substantial use of memory resources, the model makes use of orthogonal features in the form of Tchebichef moments as input. The orthogonal features have been used in achieving compression at the input stage. Considering the growing use of low energy devices, compression of neural networks becomes imperative. Here, the auto-encoder’s weights are compressed using the randomized singular value decomposition (RSVD) algorithm during training while evaluation is performed using various compression ratios. The experimental results show that the proposed fractionally compressed architecture provides improved denoising results on the standard datasets when compared with the existing methods. |
Year | DOI | Venue |
|---|---|---|
2021 | 10.1016/j.sigpro.2021.108225 | Periodicals |
Keywords | DocType | Volume |
EEG signal denoising, Auto-encoder, Fractional calculus, Orthogonal moments, Compression | Journal | 188 |
Issue | ISSN | Citations |
C | 0165-1684 | 1 |
PageRank | References | Authors |
0.36 | 0 | 3 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Subham Nagar | 1 | 1 | 0.36 |
| Ahlad Kumar | 2 | 3 | 1.39 |
| M. N. S. Swamy | 3 | 7 | 4.51 |