Skip to content

Latest commit

 

History

History
14 lines (10 loc) · 1.22 KB

week12-MultiTaskCNN.md

File metadata and controls

14 lines (10 loc) · 1.22 KB
Raw

Multi-task CNN Model for Attribute Prediction

Authors: Abrar H. Abdulnabi, Gang Wang,Jiwen Lu,and Kui Jia

Link: https://arxiv.org/pdf/1601.00400.pdf

Problem Statement & Main Contributions

  • This paper introduces a CNN framework where each CNN handles a particular task associated with an attribute. The latent representations generated by each of the CNN is combined into one shared latent feature matrix, , that allows sharing of attribute knowledge among similar groups for better prediction.

  • The main contribution of the authors is the using the implementation of the latent task matrix to generated a shared latent matrix with all the different attribute features combined and capable of sharing knowledge among similar groups.

Related Work (Key Papers)

  • The paper is heavily inspired from Zhou et.al that discusses the latent task matrix generation.

Asumptions and Future Work

  • One of the key assumptions in the paper is that sharing of knowledge amongst similar groups should be encouraged and discouraged among dissimilar groups.