=Paper=
{{Paper
|id=Vol-3288/short10
|storemode=property
|title=Autonomy of Economic Agents in Peer-to-Peer Systems (short paper)
|pdfUrl=https://ceur-ws.org/Vol-3288/short10.pdf
|volume=Vol-3288
|authors=Sergiy Obushnyi,Denis Virovets,Hennadii Hulak,Bohdan Zhurakovskyi
|dblpUrl=https://dblp.org/rec/conf/cpits/ObushnyiVHZ22
}}
==Autonomy of Economic Agents in Peer-to-Peer Systems (short paper)==
https://ceur-ws.org/Vol-3288/short10.pdf
Autonomy of Economic Agents in Peer-to-Peer Systems
Sergiy Obushnyi1, Denis Virovets1, Hennadii Hulak1, and Bohdan Zhurakovskyi2
1
Borys Grinchenko Kyiv University, 18/2 Bulvarno-Kudriavska str., Kyiv, 04053, Ukraine
2
National Technical University of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute,” 37 Peremogy ave., Kyiv,
03056, Ukraine
Abstract
The transition from a traditional economy to a digital economy based on Web 3.0 and
blockchain technologies is accompanied by some changes in the structure of relations between
participants. Such changes relate to the blurring of the concept of the ultimate beneficiary and
the center of responsibility in the case when certain digital categories are behind this or that
type of relationship, devoid of the center of control traditional for the economic system. As a
rule, such relations between participants have a high level of autonomy and a low level of
control by the state or traditional economic organizations. Thus, autonomous economical
agents, as completely independent actor, using peer-to-peer economy platforms have the
potential to have a large impact on values and behavior in society. Understanding of the
economical level of autonomy in peer-to-peer systems of such agents requires analysis of their
role in such and design of the control mechanisms in order to determine the benefits from
positive effects and at the same time mitigate negative consequences from possible mistakes.
This requires a structured overview of the levels of agent autonomy and its impact on the
existing system. The purpose of this article is to structure the study of economic agent autonomy
in peer-to-peer systems, taking into account the possibilities of the digital environment. The
article also provides an overview and analysis of the main technological developments in the
field of autonomous economic agents and decentralized autonomous organizations,
characteristics and framework of economic autonomy of the agents, taking into account digital
environment of peer-to-peer digital systems.
Keywords 1
Autonomous economic agent, Web 3.0, peer-to-peer, blockchain, DAO, decentralized
autonomous organization, P2P system.
1. Introduction replacement by a digital algorithm (Digital Twin)
[3]. The decentralization of new technologies
makes it possible to completely or partially refuse
Autonomous economic agents (AEA), as well
state protection and supervision over the activities
as Decentralized Autonomous Organizations
of such entities, while contributing to faster, safer
(DAO) [1], being designed in the peer-to-peer
and cheaper operations. The fact that digital
digital systems and acting independently in
machines (robots and computers) have proven
accordance with their internal rules represent a
their effectiveness in many areas such as finance,
new type of non-personalized (not established)
trade and banking, information storage and
subjects of economic relations described once in
analysis confirms their growing role in the digital
the works of M. Porter [2]. It is believed that in
economy, as well as their effective integration
the new decentralized (peer-to-peer) systems it
with existing economic systems.
will be difficult to determine the final
The application of blockchain technology,
personalized participant (stakeholder or
machine learning, artificial intelligence [4],
beneficiary) due to its digital anonymity, taking
digital identity, smart contracts and robotics opens
into account the possibility of its complete
up new opportunities for peer-to-peer cooperation
CPITS-2022: Cybersecurity Providing in Information and Telecommunication Systems, October 13, 2022, Kyiv, Ukraine
s.obushnyi@kubg.edu.ua (S. Obushnyi); seito@ukr.net (D. Virovets); h.hulak@kubg.edu.ua (H. Hulak); zhurakovskiybyu@tk.kpi.ua
(B. Zhurakovskyi)
0000-0001-6936-955X (S. Obushnyi); 0000-0003-4934-8377 (D. Virovets); 0000-0001-9131-9233 (H. Hulak); 0000-0003-3990-5205
(B. Zhurakovskyi)
©️ 2022 Copyright for this paper by its authors.
Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
CEUR Workshop Proceedings (CEUR-WS.org)
125
and partnership. A decentralized agent will be interaction and synchronization. Any
able to make direct peer-to-peer transactions centralization (public or private) of each of the
together with a person, or other similar digital existing modern technologies creates a number of
agent [5], which in turn makes it possible to obstacles for their optimal and sustainable
develop the idea of the economic ability of robots interaction. The creation of peer-to-peer
and bots to conclude agreements and make economic systems with elements of
transactions, where both a person and a robot can decentralization will most likely create conditions
act as a party without the necessary economic for the interaction of digital technologies and the
legal personality in the traditional sense. The emergence of a new type of economic relations
coexistence of robots and humans in the peer-to- with the participation of autonomous economic
peer systems suggests the need to study the agents. Having the ability to freely interact with
interaction between humans and robots, including each other, autonomously and securely exchange
within the framework of behavioral economics, data and digital assets, share forecasts,
law, game theory, and cryptoeconomics. autonomous economic agents will undoubtedly
Peer-to-Peer Economy Platforms are defined become a full-fledged subject of economic
in scientific papers as digital platforms where relations in the future, and, possibly, with the
providers meet directly with users without acquisition of their own separate legal status. At
intermediaries to complete a transaction with a the same time, the study of ways of interaction of
component of the physical world where there is no economic autonomous agents will be the subject
transfer of ownership [6]. This means that partici- of close study of both technical and commercial
pants enter into relationships with each other in specialists.
order to create added value using the capabilities
of peer-to-peer platforms. One such possibility is 2. Economic Autonomy of an Agent
the creation of digital autonomous agents.
Modern technologies of peer-to-peer systems in Peer-to-Peer Systems
make it possible to talk about the further
development of economic relations and the role of In a number of studies devoted to autonomous
autonomous economic agents in them with economic agents, the latter are understood as
accelerating information flows, including paired intelligent autonomous systems that act
with machine learning and artificial intelligence independently, but on behalf of and on behalf of
(AI) technologies. The possibility of achieving a users (people, participants, organizations) to solve
high level of information security, the set economic tasks within the framework of
internationalization of databases, in the conditions the granted powers. Such tasks may include
of a developed system of sensors and artificial negotiating with other agents, seeking
intelligence represent the potential for the information, interpreting past experience, and
development of the digital economy while predicting future events. Agents have mobility
optimizing a number of processes and properties; therefore, they have high performance
accelerating the development of information in dynamically distributed systems. The use of
technologies. This represents an undeniable well-designed agents in peer-to-peer systems
potential for a number of digital realms with the improves the efficiency of operations and data
increasing value of data and information as their exchange, which ultimately leads to a critical
use cases expand. reduction in transaction costs. Since autonomous
The growth of the platform business has been agents can provide intelligent services through
driven by the Internet and mobile technologies, as peer-to-peer applications, artificial intelligence
well as the rapid development of analytics, algorithms can also be successfully implemented
artificial intelligence (AI) and big data, as well as on A2A (agent to agent) platforms. At the same
changing consumer preferences and consumption time, the use of such forms of interaction is
patterns [7]. Platform business models in general, available to all traditional agents, including
and the sharing economy in particular, have led to government regulators (Fig. 1).
the creation of industries without intermediaries, To understand the role and place of an
as well as the possibility of creating autonomous autonomous agent in the economic system, we can
agents. give it the following definition: An autonomous
However, attempts to combine modern digital economic agent (AEA) is an intelligent agent
technologies in traditional systems have revealed acting on its own behalf or on behalf of the owner
a number of problems associated with their with limited intervention from the owner or other
126
agents, or without such interference, and whose that cryptographic peer-to-peer systems in their
purpose is to create economic value for its owner entirety can represent an independent intelligent
or search for its own resource. As a rule, AEAs machine.
have a narrow goal with a purposeful focus, Early autonomous agents were also presented
assuming some economic benefit. It is believed in the “Mathematical Theory of Communication”
that the autonomous operation of an agent is published in 1948 by the American electrical
achieved through the use of peer-to-peer systems engineer and mathematician Claude Elwood
and certain algorithms (smart contracts) that Shannon [9], where the author develops the topic
underlie the architecture of agents and allow of electronic communication, including with the
secure transactions without the participation of participation of independent (autonomous)
third parties. At the same time, they will be algorithms. Studying agents with their
autonomous if such a model does not require input communicative properties, the latter were
from an individual user. endowed with the following characteristics which
describe the levels of autonomy of a digital agent:
Situationality is the ability of the agent to
interact autonomously with the environment
through the use of sensors and analytical
modules.
Autonomy is the ability of an agent to
determine its actions independently without
external interference from a person or other
agents of the network.
Consistency is the ability of the agent to work
with abstract categories and draw logical
conclusions after observing and generalizing
Figure 1: Actors involved in peer-to-peer economy information.
Efficiency is the ability to perceive various
AEAs are also special in that they are created states of the environment and respond in a
to generate some economic value through timely manner to any changes.
specialized software modules or digital skills. Purposefulness is the ability of an agent to
AEA independently acquires new skills, either extract from the information flow the data
through the direct use of software modules, or necessary to implement the tasks and activate
through independent or collective learning. the appropriate algorithms, and not just
Examples of the use of AEA can be the respond to state changes, as well as the ability
acquisition of digital assets at a bargain price, to adapt to any changes in a dynamic
having the appropriate negotiation skills, while environment.
allowing the possibility of interacting with Social behavior is the ability of an agent to
another agent representing the autonomous other interact with external sources and the ability to
party to the transaction. share knowledge with other agents to jointly
solve a specific problem [10].
3. Features of Autonomy Thus, the structure of the interaction of
autonomous agents can be summarized in the
of Economic Agents following form (Fig. 2).
It is believed that the first autonomous digital
agent was a device called the Turing machine,
developed in 1948 by Alan Turing, an English
mathematician, logician and cryptographer [8].
The machine was a computing environment with
two independent agents. One agent generated
tasks, and the other solved them. Thus, the
opinion arose that agents receiving information
from the external environment can then act Figure 2: Typical building blocks of an
independently, while providing feedback and autonomous agent
communication. In addition, Turing hypothesized
127
It is believed that autonomous agents are imprinted in its internal architecture. The internal
endowed with the following properties: rationality architecture of agents and how they react to a
is an individual property of intelligent agents, as dynamic environment is highly dependent on
well as cooperative multi-agent systems or agent autonomy. Such an architecture can be
teamwork. Following the economic approach, the designed (built) and represented by abstract and
agent must maximize the utility function. To concrete classes of beliefs, desires, and intentions
study the properties of autonomous agents in (BDIs), which essentially lead to what we call
1944, von Neumann and Morgenster created mental state elements. The goal of an agent is to
Decision Theory, combining utility theory with achieve a specific set goal by following a carefully
probability theory. In decision theory, a rational crafted hierarchical plan to achieve it. An
agent is an agent that chooses an action to effective agent must have the ability to recognize
maximize expected utility, where expected utility the current situation and respond appropriately to
is defined as the actions available to the agent, the it based on their belief system. Therefore, the
probabilities of certain outcomes, and the agent's agent must be able to determine its current state in
preferences for those outcomes. In multi-agent relation to the goal being pursued.
scenarios where an agent must interact with other An autonomous agent independently makes
agents, game theory is also a powerful predictive decisions based on the conditions that the agent
and analysis tool. To solve problems with a has at its disposal. It is characteristic that the
sequence of multi-agent scenarios, in the late agent's decisions are logically limited. The beliefs
1950’s, Bellman developed Dynamic involved in decision-making are mainly related to
Programming based on the use of decision theory states (collected data about the past, present, or
methods. Particular attention was paid to the forecasts of the future, one's own skills, states, and
interoperability of agents as the ability to interact, the capabilities of other agents). The agent's
communicate and share knowledge using decisions are also constrained by previous
communication tools. decisions regarding the resources to use. For
Decentralized Autonomous Corporations example, if an agent decides to purchase
(DACs) and Decentralized Autonomous information from one database, it cannot decide to
Organizations (DAOs) are seen as forms of new purchase it from another database at the same
and innovative corporate structures that will allow time. Also, an agent cannot unilaterally revoke
new venture ideas to take root and infiltrate obligations that he has to other agents and that
business structures and have the characteristics of other agents have signed up to fulfill, but he can
an autonomous agent using blockchain cancel those obligations that other agents have to
technology and peer-2-peer systems and with a him. It is extremely important for the agent to
specific goal as to generate revenue. It is know the temporary or other criterion for
understood that such an autonomous agent exists terminating the task, otherwise he risks getting
in the cloud, performing functions that are stuck in the loop of finding the best solutions.
valuable to their owners. All operations that need As the understanding of the nature of
to be performed will be performed by the code, autonomous agents in the economic system, it
the implementation of the business logic of the became necessary to determine the place of such
DAC within the algorithm and over the an agent in the system of economic relations, as
blockchain [11]. Thus, the research of the second well as endowing him with some signs of
half of the 20th century in relation to autonomous economic subjectivity, taking into account his
agents acquires a new meaning in the context of autonomous participation in transactions. Having
peer-to-peer systems. their own structure, autonomous economic agents
The digital autonomy and independence of an act autonomously and pursue economic goals, the
agent based on peer-to-peer systems significantly achievement of which was delegated to them by a
distinguish it from other traditional participants in certain beneficiary (the owner of the agent). The
economic relations. It is believed that an autonomous agent framework facilitates user
economically independent agent should be able to experience through automation, supports
independently make decisions depending on their modularity, reuse of complex problem solutions
beliefs (modules). Therefore, the agent has and machine learning capabilities, and predicts
exclusive control over the activation of its future states that promote agent autonomy. The
services and skills, and can also refrain from use of autonomous agents is currently already
performing a task on its own. Thus, the system of available in the multi-agent peer-to-peer system
beliefs (behaviors) of an agent is arbitrarily for trading baskets of tokens [12].
128
Each agent in the real world can represent an remember and track the entire sequence of
individual, a group of people or an organization, observations. Increasing such a given database
and perform certain actions in their interests, becomes a problem for quick decision making.
maximizing economic utility. To this end, agents The reflex agent is a fairly simple agent that
must be aware of their owners' preferences and simply follows the "condition-action" rules. The
values [13]. The goal of each agent is to maximize agent perceives a certain state and acts in a certain
the outcome for their master by engaging in way, without referring to the sequences of
profitable trades based on their preferences [14]. perception. This type of agent has no autonomy at
This concept rejects the autonomous subjectivity all, because the choice of its actions is completely
of autonomous economic agents, in which agents built-in. It is possible to supplement the agent's
can achieve complete independence with algorithm with the ability to learn. The
autonomous awareness of their needs and mathematical model of the reflex agent can have
independent decision-making. We can assume the following form. The action a to perform at
that such independence may not always meet the time t + 1 can be expressed by the following state
interests of the owners of such agents. function s at time t.
Agents involved in transactions, in accordance (1)
𝑎(𝑡 + 1) = 𝑓(𝑠(𝑡)),
with their own preferences, can direct their efforts
to find strategies and a set of optimal solutions. In Stateful agents are agents that make decisions
this case, strategies may include the following: based on their internal state. The action a to be
finding suitable agents for trading; trading with performed at time t + 1 can be expressed as a
them; determining the needs of other agents to function of the expression of the state’s at time t
achieve the optimal trading sequence, etc. It is and the current internal state x(t).
believed that in this case the agent demonstrates (2)
𝑎(𝑡 + 1) = 𝑓(𝑥(𝑥(𝑡), 𝑠(𝑡))
purposeful behavior, while having the ability to
respond to state changes. From a technical point (3)
𝑥(𝑡 + 1) = 𝑔(𝑥(𝑡), 𝑠(𝑡))
of view, agents have a so-called main loop and an
event loop. The first controls the proactive Agents with an internal state can also, in turn,
behavior of the agent, in which the agent moves be classified depending on the complexity of their
towards achieving its goal at each cycle. On the algorithms into the following types:
other hand, the event loop is responsible for Deliberative agents, where the action to be
handling incoming events. Events are presented as performed is calculated based on the state of
incoming messages with their subsequent the environment, as well as taking into account
processing in the main loop [15]. the expected impact on it. In other words, the
agent motivates his actions based on the
4. Levels of Autonomy analysis of external factors.
Goal-oriented agents are agents who make
of the Economical Agents decisions given the description of desirable
situations as goals.
Depending on its functional architecture, an Utility agents are agents that can compare
economic agent may demonstrate different levels different states of the environment when
of autonomy in relation to its developer [16]. choosing a goal.
These levels are classified as follows: Planning agents are a type of more complex
Reactive agents are rather simple agents in agents that have more sophisticated built-in
their functionality, which consist only of a knowledge about the set of possible actions,
program that maps each possible sequence of understand the consequences of their actions, and
perception into the corresponding action. They also have some knowledge about the mechanisms
need built-in knowledge that uniquely defines of control of the environment. This type of agent
their behavior. They are characterized by limited is more autonomous than the previous type, since
autonomy and flexibility. They are only effective it can choose combinations of actions, but cannot
in the environment for which they were designed. be considered completely autonomous due to a
Depending on the functions of reactive agents, number of restrictions.
they are classified into a Search Agent, a Fully autonomous agents have built-in
Reflective Agent, and an Agent with an internal knowledge specific to scheduling agents and a
state. The simplest of this category of agents is the powerful learning engine. Thus, his behavior is
Search Agent. The agent uses its database to actually determined by his own experience. This
129
type of agent can define new prerequisites and operation of autonomous agents must meet the
consequences for its actions, as well as rewards following requirements: the ability to split the
for each of its actions. Examples of successful block chain to increase consistency and
learning methods are neural networks. Artificial scalability; the ability to program smart contracts
agents can use them to build and continually and develop programs compatible with the
update decision models. capabilities of machine learning and artificial
intelligence, as well as the ability to transfer these
5. Features of the Environment capabilities to other agents; an open economic
structure (OEF) embedded in an intelligent
for the Interaction database (a dynamic environment in which agents
of Autonomous Agents reside and receive input); support for fixed-point
arithmetic to ensure accuracy and determinism for
It is believed that AEA can be effectively all operations and transactions [17].
involved in the following economic areas: According to Russell and Norvig, the types of
finance; transport and logistic; supply and quality conditions for the effective existence and
control; energy market; social networks; auctions operation of independent agents are classified and
and IoT, databases and registries; personal distinguished [18], presented in table 1 below.
ratings; commercial arbitration; co-investment, Each of the types of such conditions
etc. AEAs potentially replace resellers by directly (environment) also determines the degree of its
connecting all participants in production and suitability for convenient and efficient use of the
supply chains, while reducing the need for human agent. It is believed that the most complex and
intervention and significantly saving time to meet inefficient type of environment for an agent is an
certain needs. Such a system allows multiple inaccessible, non-deterministic, dynamic and
agents to interact continuously and autonomously continuous environment. Peer-to-peer systems,
with each other without the need for any third- having different environment characteristics,
party guidance. offer different solutions and tools that can be
In order to ensure interaction between an agent attractive to AEA.
and a person, as well as autonomous agents
among themselves, including with the use of AI Table 1
technology, digital peer-to-peer ecosystems can Types of autonomous agent environments
be created with the possibility of creating and Types Characteristics
existing autonomous agents that collectively, Available and The level of information
autonomously and continuously work on solving unavailable availability in the environment
problems. At the same time, in addition to the in which the agent can receive
described characteristics, there is an opinion about
complete, accurate and up-to-
the possibility of endowing such autonomous
date information about its
agents with modular structures based on such
philosophical categories as ontology, belief, state (physical and virtual
desire, intention, abstraction, objectivity, world, the Internet).
semantics and social ability, which provides Deterministic Levels of expected guaranteed
additional advantages when interacting with a and non- results in an environment for a
person and traditional systems. deterministic particular action or set of
It is believed that the peer-to-peer environment actions and the absence of
provides the necessary level of security for the uncertainty.
operation of an autonomous agent. For an Static and The ability of an environment
autonomous agent protocol to work effectively, it dynamic to maintain its state as a result
must meet the following conditions: be stable in of the existence and activity of
the short term and unchanged in the long term; be agents within it, experiencing
scalable, which means the ability of the protocol constant changes caused by
to cope with growing and large volumes of other operations beyond the
operations, which affects the throughput of the control of individual agents.
system; be decentralized, meaning no control or
Discrete and An environment is discrete
authorization by third party groups or individuals.
continuous when it involves a fixed finite
In addition, the peer-to-peer environment for the
number of actions or calculations.
130
An environment that combines the criteria of Therefore, based on the analysis of the dynamic
security, speed and low cost of transactions will states of the environment, the agent can determine
be attractive to the user. Thus, the combination of its own behavior, referring to its goals and beliefs.
blockchain technology (peer-to-peer systems) and The use of multi-agent systems (MAS) also
agent systems opens up many opportunities for provides an opportunity for collective agent
digital partnerships, where the conditions for learning, where some autonomous agents with
interaction with other peer-to-peer platforms are competitive or mutual interests increase their
important, including the ability to build an understanding of the state and behavior in the
ecosystem of agents based on their resources. peer-to-peer ecosystems with which they are
Tools for data exchange and interaction between associated. Ideally, this will allow them to
different systems can be technologies for optimize their search for a solution to a particular
combining peer-to-peer systems, such as: problem. In turn, synergistic smart contracts (SC)
parachains; paranity; oracles, multiplexers allow developers to use the potential of the
(Multiplexer), simulators, etc. underlying blockchain infrastructure by
An ecosystem of agents can provide a system automating and executing a program or
for assessing the characteristics and states of transaction protocol in accordance with the legal
agents in order to provide system participants with (logical) terms and agreements of the contract.
information about the status of an agent and the Synergistic smart contracts are an extension of the
conditions for interacting with them. Such ratings concept of smart contracts, allowing off-grid
can arise based on the collected information about computing to be included in multi-party
agents (through the reporting module of the agreements. Such contracts allow the developer to
ecosystem), the history of their interactions with perform offline operations using machine learning
other agents, the number of positively completed models and smart databases.
tasks, as well as rating classifications and rating The presence of digital skills and abilities form
models. the basis of autonomous capabilities that AEAs
An agent operating in an environment must be can dynamically use to increase their
able to understand the various nuances of the effectiveness in various situations. The fact that an
states of such an environment in order to be able agent has one or more skills will characterize its
to predict future states. If an agent can predict the competitiveness in the ecosystem (the ability to
future, this means that he can honestly carry out work with complex tasks). Subscribing to
his actions without favoring any one action. This individual skills may depend on the strategy
concept is called the concept of justice, considered chosen by the agent. The presence of several skills
by Nassim Francez in the late 80’s [19]. The in an agent provides a system of skills priority in
ability to make predictions greatly helps this agent case of their competition. Additional skills can be
to understand the consequences of his decisions. added as packs. The ecosystem may also provide
Agents must be motivated to negotiate among for the possibility of creating various models, with
themselves in order to make the best possible the provision of access to them for individual
decisions to achieve the desired outcomes. So that agents.
agents do not get stuck on a separate process, the It is believed that digital behavior (action) is
concept of interaction provides for the priority of one or more actions, as well as their absence,
interaction in real time, the interaction of agents causing interactions with other agents initiated by
has a certain time frame, and the result must be the AEA. There are the following types of
obtained as quickly as possible, etc. behavior:
The ideal agent will be characterized by the Cyclic (CyclicBehaviour): if the agent is
ability to strike a balance between goal-directed active, the behavior remains active and is
and reactive behavior. In other words, agents must called again after each event.
be able to achieve their goal, stop pursuing the Fragmented (TickerBehaviour): a type of
goal, know when to do it—all this depends on pre- cyclic behavior in which a user-defined piece
existing environmental conditions that either of code is periodically executed)
positively or negatively affect its achievement. One-time (OneShotBehaviour): performed
All this is determined by predetermined once and self-deactivates.
conditions, such as time limits, consequences, Model (Finite State Machine or
performing or stopping the specified action. Such FSMBehaviour): a computational model that
conditions can be agreed in advance by built can be used to model sequential logic to
models, for example, real option models.
131
represent and control the execution of characteristics of autonomous and intelligent
sequential actions. In this model, fuzzy logic agents in order to take advantage of them.
can also be used to expand the range of states In fact, autonomous economics are a class of
to work with them, and using probabilities to agents with the characteristics of digital entities
determine behaviors. that can make informed and rational decisions on
Other types of agent behavior are also behalf of their stakeholders. With peer-to-peer
possible. ledger technology based on a consensus
The digital interaction module provides for the mechanism to enable secure, high-performance,
skills of synchronization with other agents, the low-cost transactions. As a result, of the
skills of negotiating and making transactions, the introduction of bridges between different types of
skills of subscribing to various protocols for peer-to-peer systems, we get a completely new
dynamically determining the states of agents, the information environment that facilitates the
skills of remembering the history of transactions introduction of autonomous agents, in which
for the purpose of subsequent training or autonomous economic agents can exist, discover
knowledge sharing, the skills of working with and be discovered, communicate with each other,
errors, etc. Thus, agents can interact for the act as an intermediary and make transactions with
purpose of jointly collecting data and information, a high level of security. The developer can use this
making available their individual skills or models environment to create agents of any caliber,
for data analysis and decision making, purpose, use, and intent. The software package for
implementing information logistics strategies or peer-to-peer systems provides tools to minimize
risk assessment, joint control of sensors, network traffic, maximize scalability and efficient
evaluating the behavior of other agents (digital use of resources. The use of agents to carry out
arbitrage), etc. commercial tasks in turn raises new questions
regarding the determination of levels of efficiency
6. Conclusions in the use of resources and the accuracy of
achieving goals. The capabilities of autonomous
agents, based on elements and tools such as
The use of independent agent technology in
beliefs, intentions, and event prediction, will
peer-to-peer systems along with artificial facilitate the use of autonomous agents in the
intelligence technology is considered fairly new. digital economy, as well as their interaction with
It is assumed that agents can be both autonomous machine learning technologies, neural networks,
and intelligent objects in the network, having a artificial intelligence, and other advanced digital
digital form in the form of a code, and reside at technologies.
the nodes, or move between them. They are
endowed with the ability to independently
identify problems or receive tasks from users or 7. References
other agents, as well as discover the necessary
resources, communicate with other agents [1] S. Obushnyi, et al., Ensuring Data Security in
(negotiate) and offer suitable solutions. They are the Peer-to-Peer Economic System of the
also good at learning from the past, updating their DAO, in Cybersecurity Providing in
knowledge, and predicting future events. The Information and Telecommunication
main differences between agents and Systems II, vol. 3187, 2021, pp. 284–292.
conventional software is the ability to [2] M. E. Porter, M. R. Kramer, How to Reinvent
independently coordinate, interact and self-learn. Capitalism—and Unleash a Wave of
By working together, agents optimally allocate Innovation and Growth.
resources, which can be like teamwork to solve a [3] S. Obushnyi, R. Kravchenko, Y. Babichen-
problem. With the ability to quickly adapt to new ko, Blockchain as a Transaction Protocol for
conditions, the use of agent technology is suitable Guaranteed Transfer of Values in Cluster
for peer-to-peer dynamic systems. Although Economic Systems with Digital Twins, in
agents have some level of dependency, they are IEEE International Scientific-Practical
endowed with communicative properties to Conference: Problems of Infocommunications
jointly search for resources necessary to solve Science and Technology, 2019, pp. 241–245.
problems. Systems designed on the basis of agent doi: 10.1109/PICST47496.2019.9061233.
technology must take into account all the [4] R. Dobre, et al., Authentication of JPEG
Images on the Blockchain, in International
132
Conference on Control, Artificial [16] N. Fornara, Interaction and Communication
Intelligence, Robotics & Optimization, 2018. among Autonomous Agents in Multiagent
[5] M. J. J. Gul, et al., Blockchain based Systems, Dissertation. https://core.ac.uk/
Healthcare System with Artificial download/pdf/20637835.pdf.
Intelligence, in International Conference on [17] F. Kipchuk, et al., Investigation of
Computational Science and Computational Availability of Wireless Access Points based
Intelligence, 2020. on Embedded Systems, in IEEE International
[6] H. J. Scholl, et al., Studying the Effects of Scientific-Practical Conference Problems of
Peer-to-Peer Sharing Economy Platforms on Infocommunications, Science and
Society, Electronic Government and Technology, 2019, pp. 246–250. doi:
Electronic, 2016. doi: 10.3233/978-1-61499- 10.1109/picst47496.2019.9061551.
670-5-222, [18] S. J. Russell, P. Norvig, Artificial
[7] J. Wirtz Platforms in the Peer-to-Peer Intelligence A Modern Approach, 1995.
Sharing Economy, Journal of Service [19] N. Francez, Fairness. Spronger-Verlag,
Management, vol. 30, no. 4, 2019, pp. 452– 1986.
483, doi: 10.1108/JOSM-11-2018-0369.
[8] A. M. Turing, Computing Machinery and
Intelligence, Mind a quarterly review of
Psychology and Philosophy, vol. LIX, no.
236, 1950.
[9] C. E. Shannon, A Mathematical Theory of
Communication, The Bell System Technical
Journal, vol. 27, 1948, pp. 379–423, 623–
656.
[10] P. G. Balaji, D. Srinivasan, An Introduction
to Multi-Agent Systems, Innovations in
Multi-Agent Systems and Applications 1.
doi: 10.1007/978-3-642-14435-6_1
[11] K. N. Kypriotaki, E. D. Zamani,
G. M. Giaglis, From Bitcoin to
Decentralized Autonomous Corporations,
Extending the Application Scope of
Decentralized Peer-to-Peer Networks and
Blockchains, in 17th International
Conferenceon Enterprise Information
Systems.
[12] D. Minarsch, et al., Autonomous Economic
Agents as a Second Layer Technology for
Blockchains: Framework Introduction and
Use-Case Demonstration, in 2020 Crypto
Valley Conference on Blockchain
Technology. doi: 10.1109/CVCBT50464.
2020.00007.
[13] K. Atkinson, T. Bench-Capon, States, Goals
and Values: Revisiting Practical Reasoning,
Argument & Computation, vol. 7, 2016, pp.
135–154. doi: 10.3233/AAC-160011.
[14] D. Minarsch, et al., Trading Agent
Competition with Autonomous Economic
Agents, Science and Technology
Publications, Lda.
[15] L. Padgham, M. Winikoff, Developing
Intelligent Agent Systems: A Practical
Guide, 2004. doi: 10.1002/0470861223.
133