| Paper ID | MLSP-20.2 |
| Paper Title |
Blind Deinterleaving of Signals in Time Series with Self-attention Based Soft Min-cost Flow Learning |
| Authors |
Oğul Can, Yeti Z. Gürbüz, Middle East Technical University, Turkey; Berkin Yıldırım, ASELSAN A. Ş., Turkey; A. Aydın Alatan, Middle East Technical University, Turkey |
| Session | MLSP-20: Attention and Autoencoder Networks |
| Location | Gather.Town |
| Session Time: | Wednesday, 09 June, 15:30 - 16:15 |
| Presentation Time: | Wednesday, 09 June, 15:30 - 16:15 |
| Presentation |
Poster
|
| Topic |
Machine Learning for Signal Processing: [MLR-SSEP] Source separation |
| IEEE Xplore Open Preview |
Click here to view in IEEE Xplore |
| Virtual Presentation |
Click here to watch in the Virtual Conference |
| Abstract |
We propose an end-to-end learning approach to address deinterleaving of patterns in time series, in particular, radar signals. We link signal clustering problem to min-cost flow as an equivalent problem once the proper costs exist. We formulate a bi-level optimization problem involving min-cost flow as a sub-problem to learn such costs from the supervised training data. We then approximate the lower level optimization problem by self-attention based neural networks and provide a trainable framework that clusters the patterns in the input as the distinct flows. We evaluate our method with extensive experiments on a large dataset with several challenging scenarios to show the efficiency. |
