Unrolled Algorithms for Group Synchronization
The group synchronization problem involves estimating a collection of group elements from noisy measurements of their pairwise ratios. This task is a key component in many computational problems, including the molecular reconstruction problem in single-particle cryo-electron microscopy (cryo-EM). Th...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Open Journal of Signal Processing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10237228/ |
| _version_ | 1797682014866898944 |
|---|---|
| author | Noam Janco Tamir Bendory |
| author_facet | Noam Janco Tamir Bendory |
| author_sort | Noam Janco |
| collection | DOAJ |
| description | The group synchronization problem involves estimating a collection of group elements from noisy measurements of their pairwise ratios. This task is a key component in many computational problems, including the molecular reconstruction problem in single-particle cryo-electron microscopy (cryo-EM). The standard methods to estimate the group elements are based on iteratively applying linear and non-linear operators, and are not necessarily optimal. Motivated by the structural similarity to deep neural networks, we adopt the concept of algorithm unrolling, where training data is used to optimize the algorithm. We design unrolled algorithms for several group synchronization instances, including synchronization over the group of 3-D rotations: the synchronization problem in cryo-EM. We also apply a similar approach to the multi-reference alignment problem. We show by numerical experiments that the unrolling strategy outperforms existing synchronization algorithms in a wide variety of scenarios. |
| first_indexed | 2024-03-11T23:54:22Z |
| format | Article |
| id | doaj.art-a70223ff64b04c39981cd7e6d26af603 |
| institution | Directory Open Access Journal |
| issn | 2644-1322 |
| language | English |
| last_indexed | 2024-03-11T23:54:22Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Open Journal of Signal Processing |
| spelling | doaj.art-a70223ff64b04c39981cd7e6d26af6032023-09-18T23:00:21ZengIEEEIEEE Open Journal of Signal Processing2644-13222023-01-01442443610.1109/OJSP.2023.331135410237228Unrolled Algorithms for Group SynchronizationNoam Janco0https://orcid.org/0000-0003-2373-9388Tamir Bendory1https://orcid.org/0000-0002-9302-0857School of Electrical Engineering, Tel Aviv University, Tel Aviv, IsraelSchool of Electrical Engineering, Tel Aviv University, Tel Aviv, IsraelThe group synchronization problem involves estimating a collection of group elements from noisy measurements of their pairwise ratios. This task is a key component in many computational problems, including the molecular reconstruction problem in single-particle cryo-electron microscopy (cryo-EM). The standard methods to estimate the group elements are based on iteratively applying linear and non-linear operators, and are not necessarily optimal. Motivated by the structural similarity to deep neural networks, we adopt the concept of algorithm unrolling, where training data is used to optimize the algorithm. We design unrolled algorithms for several group synchronization instances, including synchronization over the group of 3-D rotations: the synchronization problem in cryo-EM. We also apply a similar approach to the multi-reference alignment problem. We show by numerical experiments that the unrolling strategy outperforms existing synchronization algorithms in a wide variety of scenarios.https://ieeexplore.ieee.org/document/10237228/Algorithm unrollinggroup synchronizationmulti-reference alignment |
| spellingShingle | Noam Janco Tamir Bendory Unrolled Algorithms for Group Synchronization IEEE Open Journal of Signal Processing Algorithm unrolling group synchronization multi-reference alignment |
| title | Unrolled Algorithms for Group Synchronization |
| title_full | Unrolled Algorithms for Group Synchronization |
| title_fullStr | Unrolled Algorithms for Group Synchronization |
| title_full_unstemmed | Unrolled Algorithms for Group Synchronization |
| title_short | Unrolled Algorithms for Group Synchronization |
| title_sort | unrolled algorithms for group synchronization |
| topic | Algorithm unrolling group synchronization multi-reference alignment |
| url | https://ieeexplore.ieee.org/document/10237228/ |
| work_keys_str_mv | AT noamjanco unrolledalgorithmsforgroupsynchronization AT tamirbendory unrolledalgorithmsforgroupsynchronization |