التفاصيل البيبلوغرافية
العنوان: |
Reconstruction of data modes per expert. |
المؤلفون: |
Andreas Kopf (3741847), Vincent Fortuin (10765742), Vignesh Ram Somnath (11054112), Manfred Claassen (2697070) |
سنة النشر: |
2021 |
المجموعة: |
Smithsonian Institution: Digital Repository |
مصطلحات موضوعية: |
Genetics, Evolutionary Biology, Immunology, Hematology, Infectious Diseases, Biological Sciences not elsewhere classified, Mathematical Sciences not elsewhere classified, Chemical Sciences not elsewhere classified, Information Systems not elsewhere classified, MoE-Sim-VAE exhibits, multi-modal distributions, Mixture-of-Experts Similarity Varia., Deep Clustering, high-dimensional data, Gaussian mixture distribution, mouse organs, data points, Mixture-of-Experts Variational Auto., MNIST benchmark data, cell data Clustering high-dimension., cell subpopulations, novel generative, RNA-sequencing measurements, competitor methods, mass cytometry, model, similarity-based representations, Variational Autoencoder |
الوصف: |
PCA plot showing the reconstruction (red) of original data (colored underneath) separated per MoE-expert on the Inhibitor GDC-0941 and Well A09 from the Bodenmiller [ 47 ] data. This example reached a F-measure of 0.8606. The experts with ID 2, 3, …, 9 where not selected via the gating network. The red samples in each plot visualize the reconstructed data. A) Expert ID = 0. B) Expert ID = 1. C) Expert ID = 10. D) Expert ID = 11. E) Expert ID = 12. F) Expert ID = 13. G) Expert ID = 14. H) Visualization of the reconstruction taking the data modes from all selected experts together. I) PCA plot of the true labels without any reconstruction overlaid. (EPS) |
نوع الوثيقة: |
still image |
اللغة: |
unknown |
Relation: |
https://figshare.com/articles/figure/Reconstruction_of_data_modes_per_expert_/14887770 |
DOI: |
10.1371/journal.pcbi.1009086.s009 |
الاتاحة: |
https://doi.org/10.1371/journal.pcbi.1009086.s009 |
Rights: |
CC BY 4.0 |
رقم الانضمام: |
edsbas.53A5D702 |
قاعدة البيانات: |
BASE |