An Efficient Iterative Algorithm to Explainable Feature Learning

Dino Vlahek

Abstract


This paper summarizes a doctoral thesis introducing the new iterative approach to explainable feature learning. Features are learned in three steps during each iteration: feature construction, evaluation, and selection. We demonstrated superior performances compared to the state of the art on 13 of 15 test cases and the explainability of the learned feature representation for knowledge discovery.

Full Text:

PDF

References


Y. Bengio, A. Courville, and P. Vincent. Representation Learning: A Review and New Perspectives. IEEE Transactions on Pattern Analysis and Machine Intelligence, 35(8):1798–1828, aug 2013.

Mehrdad J. Gangeh, Ahmed K. Farahat, Ali Ghodsi, and Mohamed S. Kamel. Supervised dictionary learning and sparse representation-a review. ArXiv, abs/1502.05928, 2015.

Huan Liu and Hiroshi Motoda. Feature Ex-

traction, Construction and Selection: A Data

Mining Perspective. Kluwer Academic Publishers, Norwell, MA, USA, 1998.

Yunqian Ma and Yun Fu. Manifold Learning Theory and Applications. CRC Press, Inc., USA, 1st edition, 2011.

Michael A. Nielsen. Neural Networks and Deep Learning, page 216. Determination Press, 2018.

W. Tang, A. Panahi, H. Krim, and L. Dai. Analysis dictionary learning based classification: Structure for robustness. IEEE Transactions on Image Processing, 28(12):6035–6046, Dec 2019.

Dino Vlahek. Učinkovit iterativni algoritem učenja razložljivih značilnic za izboljšano klasifikacijo. Ph.D . disertation, UM FERI, 2024.




DOI: https://doi.org/10.31449/inf.v48i2.6105

Creative Commons License
This work is licensed under a Creative Commons Attribution 3.0 License.