BZPEERReading, trust and collaboration for research
Menu

Discover

GraphFollow connections around a paper, topic or saved view.

Workspaces

HomePick up where your research flow left off.InboxHandle requests, replies and notifications from one place.CollectionsCurate reading lists, evidence sets and shareable dossiers.ResearchSee your private provenance graph, trail and imports.CollaborationMove from discovery into focused discussion rooms.PublishDraft, preview and publish full-text research articles.

Network

ExpertsFind specialists worth following or asking for help.CommunitiesJoin research circles organized around shared work.PartnersSee external organizations active around the same topics.

Opportunities

ListingsBrowse roles, calls and collaborations with clearer fit signals.

Account

Log inReturn to your reading and collaboration workspace.Create accountStart saving work, following people and joining teams.
Log inCreate account

Researcher profile

Mojtaba Moattari

Mojtaba Moattari contributes to research discovery and scholarly infrastructure.

ResearcherAffiliation not importedOpen to collaborate
Neural and Evolutionary ComputingArtificial IntelligenceMachine Learning

Trust snapshot

Quick read

Trust 15 - Baseline
3works
0followers
3topics
4close collaborators

Actions

Decide how to stay connected

Follow researcher0

Research graph

See the researcher in context

Open full explorer

Inspect adjacent work, topics, institutions and collaborators without jumping out to a separate graph page.

Building this graph slice

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Published work

3 published item(s)

preprint2020arXiv

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

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 equations interacting together. Such search spaces are provided with multiple convex regions peaked with diverse nonlinear slopes and in non-homogenous ways. To find global optima, a new meta-heuristic algorithm is proposed based on Observer Effect concepts for controlling memory usage per localities without pursuing Tabu-like cut-off approaches. Observer effect in physics (or psychology) regards bias in measurement (or perception) due to the interference of instrument (or knowledge). Performance analysis of the proposed algorithms is sought in two real-world engineering applications, i.e., Electroencephalogram feature learning and Distributed Generator parameter tuning, each of which having nonlinearity and complex multi-modal peaks distributions as their characteristics. In addition, the effect of version improvement has been assessed. The performance comparison with other optimizers in the same context suggests that proposed algorithm is useful both solely and in hybrid Gradient Descent settings where problem's search space is nonhomogeneous in terms of local peaks density.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

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

Gradient Descent (GD) approximators often fail in the solution space with multiple scales of convexities, i.e., in subspace learning and neural network scenarios. To handle that, one solution is to run GD multiple times from different randomized initial states and select the best solution over all experiments. However, this idea is proved impractical in plenty of cases. Even Swarm-based optimizers like Particle Swarm Optimization (PSO) or Imperialistic Competitive Algorithm (ICA), as commonly used GD initializers, have failed to find optimal solutions in some applications. In this paper, Swarm-based optimizers like ICA and PSO are modified by a new optimization framework to improve GD optimization performance. This improvement is for applications with high number of convex localities in multiple scales. Performance of the proposed method is analyzed in a nonlinear subspace filtering objective function over EEG data. The proposed metaheuristic outperforms commonly used baseline optimizers as GD initializers in both the EEG classification accuracy and EEG loss function fitness. The optimizers have been also compared to each other in some of CEC 2014 benchmark functions, where again our method outperforms other algorithms.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Uncertainty Principle based optimization; new metaheuristics framework

To more flexibly balance between exploration and exploitation, a new meta-heuristic method based on Uncertainty Principle concepts is proposed in this paper. UP is is proved effective in multiple branches of science. In the branch of quantum mechanics, canonically conjugate observables such as position and momentum cannot both be distinctly determined in any quantum state. In the same manner, the branch of Spectral filtering design implies that a nonzero function and its Fourier transform cannot both be sharply localized. After delving into such concepts on Uncertainty Principle and their variations in quantum physics, Fourier analysis, and wavelet design, the proposed framework is described in terms of algorithm and flowchart. Our proposed optimizer's idea is based on an inherent uncertainty in performing local search versus global solution search. A set of compatible metrics for each part of the framework is proposed to derive preferred form of algorithm. Evaluations and comparisons at the end of paper show competency and distinct capability of the algorithm over some of the well-known and recently proposed metaheuristics.

0 citations0 reviewsNo code linkedOpen access