Multi-view label distributed learning with multiple sub-networks
–>
Multi-view label distribution learning with multiple sub-networks

Zhang HengRu1[00000 11111 22222 3333], Rong
–>
Bin-Yuan Rong, Heng-Ru Zhang

Label distribution learning is an emerging learning paradigm to address label ambiguity, which can describe the degree of correlation between different labels and sample.
–>
Label distribution learning (LDL) is an emerging learning paradigm to address label ambiguity, which can describe the degree of correlation between labels and samples.

However, it has not yet involved the use of multiple views to obtain diverse representations and rich semantics.
–>
However, there is no related work involving obtaining different representations and rich semantics of samples through multiple views.

In this paper, we propose an end-to-end deep learning model called MSNN-LDL, which uses multiple sub-networks to extract semantic features from different views and improve the robustness of the model.
–>
In this paper, we propose an end-to-end deep learning model called MSN-LDL, which uses Multiple Sub-Networks to simulate the semantic features of samples extracted from different views.

Keywords: Label distribution learning · Label ambiguity · Deep learning · Neural networks.
–>
Keywords: Deep learning · Label distribution learning · Multiple views · Multiple sub-networks.