Separation of VLC signals using FastIca and InfoMax

E San Juan, I Soto, G Salinas… - 2017 First South …, 2017 - ieeexplore.ieee.org
This paper proposes two methods of separation of VLC signals, based on the stadistical
variation 1) FastIca and 2) InfoMax, the first one based in the stadistical covariance of the …

Real-time framework for human action recognition

HH Ali, HM Moftah, AAA Youssif - Proceedings of the 2018 8th …, 2018 - dl.acm.org
This paper introduces a real-time framework for human action recognition by Microsoft
sensor for Windows. The framework detects human skeleton joints and extracts human from …

[PDF][PDF] Empowering swarm-based optimizers by multi-scale search to enhance Gradient Descent initialization performance.

M Moattari, MH Moradi, R Boostani - CoRR, 2019 - academia.edu
Empowering swarm-based optimizers by multi-scale search to enhance Gradient Descent
initialization performance Page 1 Empowering swarm-based optimizers by multi-scale …

A New Approach for Optimizing Highly Nonlinear Problems Based on the Observer Effect Concept

M Moattari, E Roshandel, S Kamyab… - arXiv preprint arXiv …, 2019 - arxiv.org
A lot of real-world engineering problems represent dynamicity with nests of nonlinearities
due to highly complex network of exponential functions or large number of differential …

Modified swarm-based metaheuristics enhance Gradient Descent initialization performance: Application for EEG spatial filtering

M Moattari, MH Moradi, R Boostani - arXiv preprint arXiv:1907.08220, 2019 - arxiv.org
Gradient Descent (GD) approximators often fail in the solution space with multiple scales of
convexities, ie, in subspace learning and neural network scenarios. To handle that, one …