Action unknown: copypageplugin__copy

10a.002.TAU_WP6 - Operational Support Estimator Network (OSEN) for Sparse Support Estimation and Learning-Aided Compressive Sensing

Project - Summary

  • An approach for energy efficient Support Estimation (SE) and learning-aided Compressive Sensing (CS).
  • Non-iterative SE is accomplished by the proposed Operational Support Estimator Networks (OSENs).
  • Highly improved non-linearity compared to traditional convolutional layers.
  • OSENs are designed with “generative super-neurons” with non-localized kernels.
  • This project builds the first step of solving linear inverse problems using networks with non-linear neuron models.

Project - Team

Team Member Role Email Phone Number Academic Site/IAB
Moncef Gabbouj PI 358 40 073 6613 Tampere University
Serkan Kiranyaz Co-PI 97 43 063 5600 Tampere University
Mete Ahishali Researcher 358 46 552 3736 Tampere University
Matti Vakkuri Project Mentor N/A Haltian

Project - Novelty of Approach

See “Project Summary” section above.

Project - Deliverables

1 Improved classification framework with the hybrid loss penalizing both SE estimation and classification errors during training.
2 Learning-aided CS scheme with the prior information produced by the OSEN.

Project - Benefits to IAB

  • Advanced Machine Learning especially for scarce data is a great asset for any service company offering AI solutions to real world problems.
  • The proposed approach can be used in many tasks such as classification, anomaly localization, and various CS based applications including compressive video surveillance systems and Distributed Compressive Sensing based surveillance.
  • Learning-aided CS scheme: the output of OSEM also provides the probability of support sets that will be used as prior in the classical CS schemes. In this way, the convergence time of the classical methods are drastically reduced.

Project - Documents

projects/year10/10a.002.tau_wp6.txt · Last modified: 2022/05/13 08:27 by sally.johnson