China
Africa
Explore chapters and articles related to this topic
Algorithms and Tools
Published in Hamidreza Ahmadian, Roman Obermaisser, Jon Perez, Distributed Real-Time Architecture for Mixed-Criticality Systems, 2018
Most exploration algorithms are based on generic optimization or constraint solving algorithms, often including custom algorithmic extensions and rarely use distinct search algorithms [147]. In [148], a SMT solver is used to obtain an initial solution for an Evolutionary Algorithm (EA) to optimize a deployment of software components to a hardware platform. For the chosen case study, that has a large and restrictive constraint set that is modeled using a design specific language, the initial solution was not further optimized by the EA. The meta-optimization framework Opt4J [152] includes solvers based on a genetic algorithm, SAT formulations, and others. It supports multi-objective optimization by means of a Pareto-Front and has been primarily applied to task allocation problems, e.g.,[153]. Beyond the scope of the task allocation problem, [149] also explores the set of optimal hardware platform architectures by a step-wise increase of the platform’s granularity and instantiating hardware elements by elements from a library using a framework incorporating different solver methods, e.g.,mixed-integer linear programming.
Deep Learning: Parameter Optimization Using Proposed Novel Hybrid Bees Bayesian Convolutional Neural Network
Published in Applied Artificial Intelligence, 2022
Nawaf Mohammad H Alamri, Michael Packianather, Samuel Bigot
Chiroma et al. (2019) discussed the synergy between nature inspired algorithms with DL. They mentioned that the inspiration of such algorithms can be from animals’ behaviors, human activities, and biological systems. The paper presented many nature inspired algorithms such as harmony search, firefly, cuckoo search, evolutionary, ant colony optimization, practical swarm optimization, genetic, simulated annealing, and gravitational search algorithm. The authors stated that combining DL with nature inspired algorithms has the advantage of solving local minima problem and improving the performance of the network by increasing the accuracy of its architecture. In addition, the need for trial-and-error techniques in determining the parameters of DL architecture is eliminated as nature inspired algorithms realize the best parameters values automatically. Although the optimum parameters setting is still an open problem in the research area. The authors suggested to eliminate the need for human interventions in determining the parameters by obtaining parameter-less nature inspired algorithms in the future. Finally, the paper suggested to apply meta-optimization that is excessive in the DL area, and it helps to tune optimization methods by using another optimization method.