این رویداد خاتمه یافته است و اطلاعات موجود در این سایت صرفا جنبه آرشیو دارد

:: چکیده سخنرانی



Data-Driven Fuzzy Modeling
Irina Perfilieva, University of
Ostrava
 
The talk will focus on efficient data-driven modeling associated with the inverse problem and feature extraction. We show how the theories of manifolds and F-transforms contribute to these delineated areas.
 
The manifold hypothesis states that the shape of observed data is relatively simple and that it lies on a low-dimensional manifold embedded in a higher-dimensional space.
We contribute to the problem of manifold learning. We show that a space whose topological structure is characterized by a fuzzy partition naturally leads to so called Riemannian spaces or Riemannian manifolds.
 
Finally, we show how the discussed notions contribute to the mathematics of deep learning.
 

نمایه IEEE





فایل‌های مورد نیاز

خبر تصویری

1
2
3
4
5
6
7
8
9

چاپ مقالات منتخب در

پوستر همایش

© کلیه حقوق این وب سایت محفوظ می باشد .
طراحی و پیاده سازی شده توسط : همایش نگار ( ویرایش 10.0.5)