Mathematical support and software for data processing in robotic neurocomputer systems

Romanchuk, V. A.

doi:10.1051/matecconf/201816103004

Cited by 2 publications

(1 citation statement)

References 10 publications

(13 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Neurocomputers are computers of the 6th generation ( Figure 1) and are considered more efficient computing devices, but only when solving problems presented in a neural network logical basis. In this regard, many companies are actively engaged in the production of neuroprocessor technology: Google, NVidia, Qualcomm, IBM, Toshiba, HumanBrainProject, KnuEdge, AnalogDevices, Texas Instruments, Darwin, Fujitsu, Eyeriss, Intel and others [16]. In 2013, Qualcomm introduced its prototype Zeroth processor, which simulates the properties of the human brain's, including the property of self-learning in the process.…”

Section: Application Of Hardware Of Artificial Intelligence Systementioning

confidence: 99%

Artificial Intelligence Technologies in Digital Modernization of Organizations

Romanchuk¹,

Kartashov²,

Tokhtieva³

et al. 2020

Proceedings of the Russian Conference on Digital Economy and Knowledge Management (RuDEcK 2020)

View full text Add to dashboard Cite

Currently, people in different media sources is actively discussing the negative effects of computers and robotic systems that can affect various aspects of people's lives, significantly limiting it. This article describes the terms: "machine learning", "artificial intelligence", "neural networks"; some prerequisites and possibilities of using artificial intelligence systems in various areas of society are considered. The detailed analysis of the necessary computing devices for solving the problems of digital transformation of organizations is presented. Concrete examples of the use of artificial intelligence systems in modern sectors of the economy, at enterprises and in institutions. This paper gives examples of the use of decision support systems based on artificial intelligence technologies in various fields. As possible computing devices for accelerating neural network operations, graphics cards, specialized neural network accelerators, and neuromorphic processors are considered. The article notes that the architecture of neural network computers is more promising than the classical one Neumann architectures currently used due to the possibility of highly parallel operation and very low energy consumption in the process of solving problems. As a result of the analysis of the technologies used, the authors substantiate the conclusion that a neural network is software and ideally suited for solving the issues of digital modernization of an organization. A neural network, learning from specific examples given by humans, is able clearly fulfill the tasks set before it and, if necessary, make decisions independently. With the introduction of artificial intelligence systems in the organization, there will be significant savings in material and labor resources as a result of reduced decisionmaking time.

show abstract

Section: Application Of Hardware Of Artificial Intelligence Systementioning

confidence: 99%