Are Direct Links Necessary in Random Vector Functional Link Networks for Regression?

This is a video presentation for the paper "Are Direct Links Necessary in Random Vector Functional Link Networks for Regression?" The work was presented at the 19th International Conference on Artificial Intelligence and Soft Computing, October 12-14, 2020.
The paper:
arxiv.org/abs/..., link.springer.....
Abstract:
A random vector functional link network (RVFL) is widely used as a universal approximator for classification and regression problems. The big advantage of RVFL is fast training without backpropagation. This is because the weights and biases of hidden nodes are selected randomly and stay untrained. Recently, alternative architectures with randomized learning are developed which differ from RVFL in that they have no direct links and a bias term in the output layer. In this study, we investigate the effect of direct links and output node bias on the regression performance of RVFL. For generating random parameters of hidden nodes we use the classical method and two new methods recently proposed in the literature. We test the RVFL performance on several function approximation problems with target functions of different nature: nonlinear, nonlinear with strong fluctuations, nonlinear with linear component and linear. Surprisingly, we found that the direct links and output node bias do not play an important role in improving RVFL accuracy for typical nonlinear regression problems.
Keywords:
Random vector functional link network, Neural networks with random hidden nodes, Randomized learning algorithms.

[@kohzhewei9677]

Thank you for the video!

[@juancamiloalzate1481]

Thank you for the video. It helped me a lot

[@lingyi247]

Hi, Professor Dudek. Thank you for the presentation. However, it makes me confused that I have read another paper, , which presents the opposite results. They said, " the direct link plays an important performance enhancing role in RVFL, while the bias term in the output neuron had no significant effect". Can you explain it? Thank you so much!

[@grzegorzdudek2236]

Hi Ling Yi. Note that I used RVFL for regression problems, while in the paper you cite they focus on classification problems. This is a substantial difference. Another difference is the method of generating random parameters. I use my own methods described in Section 3 arxiv.org/abs/2003.13090. Perform your own tests to see if direct links are necessary :)

[@lingyi247]

@@grzegorzdudek2236 Thank you for your prompt reply. That really helped! :)