Metagraph Representation for Overcoming Limitations of Existing Knowledge Bases

The paper describes an approach to using mivar technologies for planning three-dimensional robot routes, taking into account the given obstacles. This system is of great importance in increasing the autonomy of robots, as it will ultimately help us get closer to creating artificial intelligence in mechanical engineering. At present, there are many tasks related to automation and robotics that require non-trivial solutions. The article describes three models of three-dimensional logical space and presents a visualization of route construction in these spaces for robotic complexes. Obstacles in three-dimensional logical space are understood as the absence of graph vertices and transitions between them in a certain specific area. The use of mivar technologies of logical artificial intelligence in the field of transport systems allows to significantly speed up route planning thanks to a unique mivar algorithm for processing information, which allows creating systems capable of making decisions in real time. This work is intended for researchers dealing with the problem of three-dimensional route planning using mivar technologies.

Section: Introductionmentioning

confidence: 99%

Route planning of autonomous robots in three-dimensional logic space using mivar technologies

Kotsenko,

Andreev,

Kim

et al. 2024

“…HIIS can be used, for example, for generating advertising videos [10] or for analysis [11] and creating a brief overview [12] of judicial practice. In addition to the mivar approach, metagraph representation [13] of datasets [14] can be used in HIIS to overcome the limitations [15] of existing knowledge bases [16].…”

Section: Introductionmentioning

confidence: 99%

Automated search for several alternative logical inferences in a mivar knowledge base

Kotsenko,

Bazanova,

Kim

et al. 2024

This paper describes an approach to the automatic generation of a mivar knowledge base of a three-dimensional logic space. This knowledge base is created to provide three-dimensional robot movement. Automatic generation of mivar knowledge bases is necessary for the further solution of the problem of planning three-dimensional routes of robots and robotic complexes. This work is the basis for solving problems of resource allocation optimization in the field of transport logistics and machine learning artificial intelligence based on the application of mivar technologies. The variant of practical realization of the automated technique of sequential removal of mivar network rules and search for new robot motion trajectories for their further comparison by the number of activated transition rules on the way from the initial location to the target one is considered. Mivar-based logic artificial intelligence technologies offer great opportunities for dynamic route finding in real time, which will help in further development of machine learning artificial intelligence in the field of transport systems. The paper is intended for researchers involved in the development of mivar expert systems for solving problems of three-dimensional movement of robots and optimizing production planning systems.

Section: Introductionmentioning

confidence: 99%

Evaluation of BERT and ChatGPT models in inference, paraphrase and similarity tasks

Kim,

Kotsenko,

Andreev

et al. 2024

The purpose of this paper is to study the application of ChatGPT and BERT models in the field of mechanical engineering. In the context of machine learning, the ChatGPT and BERT models can be applied to various natural language processing tasks such as analyzing technical documentation and building instructions according to a particular version of the documentation, diagnosing malfunctions or customer service. The paper discusses the fundamental features of BERT and ChatGPT models, their origin, and also investigates the main architectural features and identifies the main advantages and disadvantages of the models. The paper analyzes and selects various natural language processing tasks to test the models’ ability to understand natural language in the context of machine learning. The selected criterion tasks are divided into semantic groups to identify the capabilities of ChatGPT and BERT models in each of three areas: logical inference tasks, paraphrasing tasks, and text similarity tasks. The paper also discusses the concept of operational design, which involves developing inputs that guide the models to produce desired outputs. The paper quantitatively analyzes and compares the performance of BERT and ChatGPT based models. The reasons for the bottlenecks of ChatGPT model in natural language understanding tasks are discovered and investigated. Possible improvements of ChatGPT model performance using the mivar approach are considered.