Artificial Intelligence Problems Research Articles

Civil law regime of objects generated by artificial intelligence

The article is devoted to the civil law aspects of the problems of artificial intelligence and its place in the structure of civil law relations, as well as questions regarding the legal regime of objects generated by artificial intelligence. The approaches expressed in the domestic legal doctrine regarding the understanding of artificial intelligence as an object and as a subject of civil law relations, as well as a point of view regarding the possibility of simultaneously considering artificial intelligence as an object and as a subject of such legal relations, are analyzed. On the basis of approaches to the definition of the concept of «artificial intelligence», which are beginning to be formed at the level of legal acts of the United Nations and the European Union, as well as enshrined in the legislation of Ukraine, it is concluded that artificial intelligence must be considered as an object of civil law relations. In this regard, a critical position has been expressed regarding attempts made at the theoretical level to interpret artificial intelligence or robots endowed with artificial intelligence as subjects equivalent to natural persons, as quasi-legal entities. The main provisions of the Law of Ukraine «On Copyright and Related Rights» on the sui generis right on a non-original object generated by a computer program, which can be considered corresponding concept of «object generated by artificial intelligence», have been analyzed. The features of a non-original object generated by a computer program are singled out as a special type of an object of sui generis right. A position has been expressed regarding the expediency of clarifying the provisions of the Law of Ukraine «On Copyright and Related Rights» in the part of defining the subject of a sui generis right to a non-original object generated by a computer program. It was concluded that objects generated by artificial intelligence can be classified as objects of intellectual property law, within the framework of which their legal protection should be ensured by a special kind of right – sui generis right. The expediency of enshrining the main provisions regarding the sui generis right to an object generated by artificial intelligence in Book 4 of the Civil Code of Ukraine is emphasized.

In Search of Dispersed Memories: Generative Diffusion Models Are Associative Memory Networks.

Uncovering the mechanisms behind long-term memory is one of the most fascinating open problems in neuroscience and artificial intelligence. Artificial associative memory networks have been used to formalize important aspects of biological memory. Generative diffusion models are a type of generative machine learning techniques that have shown great performance in many tasks. Similar to associative memory systems, these networks define a dynamical system that converges to a set of target states. In this work, we show that generative diffusion models can be interpreted as energy-based models and that, when trained on discrete patterns, their energy function is (asymptotically) identical to that of modern Hopfield networks. This equivalence allows us to interpret the supervised training of diffusion models as a synaptic learning process that encodes the associative dynamics of a modern Hopfield network in the weight structure of a deep neural network. Leveraging this connection, we formulate a generalized framework for understanding the formation of long-term memory, where creative generation and memory recall can be seen as parts of a unified continuum.

인공지능(AI)과 한국 부동산의 부패방지에 관한 연구

This study attempted to present discussions and directions regarding artificial intelligence(AI) and the prevention of corruption in Korean real estate. This Paper dealt with moral hazard, corruption prevention, and trust using real estate, the 4th industrial revolution(4IR), AI, and intelligent information technology as keywords. Prior research is focusing on economic utility and technical benefits related to AI and real estate. In contrast, this study pointed out the risk of moral hazard and corruption arising from the combination of AI and real estate, and conducted a paper to prevent moral hazard and corruption. In this respect, there is a difference between this study and existing previous studies. This is where the originality of this study appears. The results of this study on anti-corruption of AI and Korean real estate are as follows. As a result of the study, the problems of AI and Korean real estate, and anti-corruption were presented in three major categories: 1. AI and the problem of changing times, 2. Korean real estate and corruption problem, and 3. AI and real estate problem. In relation to this, this study largely derived three measures as implications of the research. The contents include, first, responses to the issues of AI and changing times, second, responses to Korean real estate and corruption issues, and third, measures to respond to AI and real estate issues. This study hopes to restore market trust through Artificial Intelligence(AI) and the prevention of corruption in Korean real estate, and to build a future where the benefits and utility provided by real estate, AI, and intelligent information technology are desirable.

DYNAMIC PROGRAMMING FOR SOLVE THE INVENTORY MANAGEMENT PROBLEM IN LOGISTICS

Currently, there is a problem of methods insufficient efficiency for finding solutions to the inventory management problem. The research object is the process of solving inventory management problems. The research subject is methods for finding a solution to the inventory management problem based on dynamic programming. The research goal is to increase the efficiency of finding a solution to the inventory management problem through dynamic programming. A method based on deterministic dynamic programming, a method based on stochastic dynamic programming, a method based on Q-learning, and a method based on SARSA were applicated for the inventory management problem. There are advantages of the methods. of Methods modification of deterministic and stochastic dynamic programming, Q-learning, and SARSA due to dynamic parameters makes it possible to increase the learning speed while maintaining the root-mean-square error of the method. The numerical study made it possible to evaluate the methods (for modifying the deterministic and stochastic dynamic programming methods, the number of iterations is close to the number of stages; for both methods of deterministic and stochastic dynamic programming, the root mean square error was 0.02; for modifying the Q-learning and SARSA methods, the number of iterations was 300, for both methods of Q-learning and SARSA, the root mean square error was 0.05). These methods make it possible to expand the scope of dynamic programming, which is confirmed by their adaptation to the inventory management problem and helps to increase the intelligent computer systems efficiency for general and special purposes. The application of these methods for a wide class of artificial intelligence problems are the prospects for further research.

God from the machine 2.0: biblical narrative as a metalanguage for discussing machine ethics

The article is devoted to theoretical models in which the biblical narrative becomes a means of conceptualizing the actions of technological objects. Based on the works of Russian and foreign philosophers, it is shown what the specifics of the discursive formulation of the problems of machine ethics, understood as a subject, are. The focus is not on the formulation of the problem of artificial intelligence ethics per se or the possibility of solving it, but on the very method of discursive formulation, or the way of talking about it, by incorporating biblical plots, motifs, and images as mechanisms for understanding this problem. The subject of the study is the heuristic potential of the biblical metanarrative in the structure of statements about artificial intelligence. It is demonstrated that the heuristic potential of the biblical metanarrative in the study of AI problems is related to the production of distinctions. It is revealed which new systems of basic distinctions can be constructed using this method of discursive design. It is shown what changes are observed in the discourse about algorithmic technologies. Possible reasons for the activation of biblical discourse in addressing the problems of artificial intelligence are identified.

Identification for Tree-Shaped Structural Causal Models in Polynomial Time

Linear structural causal models (SCMs) are used to express and analyze the relationships between random variables. Direct causal effects are represented as directed edges and confounding factors as bidirected edges. Identifying the causal parameters from correlations between the nodes is an open problem in artificial intelligence. In this paper, we study SCMs whose directed component forms a tree. Van der Zander et al. give a PSPACE-algorithm for the identification problem in this case, which is a significant improvement over the general Gröbner basis approach, which has doubly-exponential time complexity in the number of structural parameters. However, they do not show that their algorithm is complete. In this work, we present a randomized polynomial-time algorithm, which solves the identification problem for tree-shaped SCMs. For every structural parameter, our algorithms decides whether it is generically identifiable, generically 2-identifiable, or generically unidentifiable. (No other cases can occur.) In the first two cases, it provides one or two fractional affine square root terms of polynomials (FASTPs) for the corresponding parameter, respectively. In particular, our algorithm is not only polynomial time, but also complete for for tree-shaped SCMs.

ARTIFICIAL INTELLIGENCE AS A BASIC PROBLEM WHEN IMPLEMENTING AUTONOMOUS VEHICLE TECHNOLOGY IN EVERYDAY LIFE

Innovative technologies that use artificial intelligence in transport solutions recently emerging around the world include, among others issues of autonomous vehicle driving. The use of autonomous vehicle technology affects the issues of civil liability (liability and insurance), road safety, natural environment (energy efficiency, renewable energy sources), data (access, exchange, protection, privacy), IT infrastructure (effective and reliable communication), employment (creation and loss of jobs, training of truck drivers in the use of automated vehicles). The development of new technologies related to artificial intelligence, including autonomous vehicles, generates inevitable changes in law, economy and society. It is inevitable due to the fact that autonomy is undoubtedly a means to achieve the goal of improving the efficiency sought in every area of life. The article presents arguments confirming the thesis that the basic factor inhibiting the implementation of autonomous vehicle technology is the problem of artificial intelligence, including its definition and legal regulation.

Motivations for Artificial Intelligence, for Deep Learning, for ALife: Mortality and Existential Risk.

We survey the general trajectory of artificial intelligence (AI) over the last century, in the context of influences from Artificial Life. With a broad brush, we can divide technical approaches to solving AI problems into two camps: GOFAIstic (or computationally inspired) or cybernetic (or ALife inspired). The latter approach has enabled advances in deep learning and the astonishing AI advances we see today-bringing immense benefits but also societal risks. There is a similar divide, regrettably unrecognized, over the very way that such AI problems have been framed. To date, this has been overwhelmingly GOFAIstic, meaning that tools for humans to use have been developed; they have no agency or motivations of their own. We explore the implications of this for concerns about existential risk for humans of the "robots taking over. " The risks may be blamed exclusively on human users-the robots could not care less.

Artificial Intelligence in the Context of Criminal Law Risks

Relevance. The growing digitalization of society, leads to the emergence of new forms of socially dangerous behavior (socially dangerous activity). In many ways, such criminal and legal risks are caused by the involvement of the phenomenon of artificial intelligence in various spheres of human activity. During the operation of artificial intelligence, it is also possible to exert harmful effects on artificial intelligence itself and (or) its carrier, which, from the standpoint of the current criminal law, may not always receive an unambiguous qualification.The purpose of the study is to identify criminal law risks existing in the context of artificial intelligence and formulate scientifically based conclusions regarding the prospects for the development of domestic criminal law and legislation.Objectives: to identify the key criminal law risks associated with the exploitation of artificial intelligence; to check whether artificial intelligence has the properties necessary for criminal personality; to identify possible options for criminal law assessment of harm caused by the exploitation of artificial intelligence; to establish the sufficiency of criminal law resources to protect artificial intelligence itself from socially dangerous behavior.Methodology. The methodological basis of the research was the universal dialectical method of cognition of phenomena and processes of the surrounding reality. During the development of theoretical and applied provisions of the work, a set of general scientific and private scientific research methods (formal logical, predictive, formal legal, etc.) was also used.Results. The prospects for the development of criminal law and criminal legislation of Russia in the context of the problem of artificial intelligence directly depend on the level of scientific and technological achievements in its programming and operation.Conclusion. The modern potential of artificial intelligence precludes raising the question of its criminal legal personality mainly due to its lack of ability to mentally perceive its own socially dangerous activity, which is a prerequisite for criminal liability. The developer of the corresponding program or the operator of the artificial intelligence carrier device must be responsible for causing harm to interests protected by criminal law in connection with the exploitation of artificial intelligence.

A feeling for the algorithm: Diversity, expertise, and artificial intelligence

Diversity is often announced as a solution to ethical problems in artificial intelligence (AI), but what exactly is meant by diversity and how it can solve those problems is seldom spelled out. This lack of clarity is one hurdle to motivating diversity in AI. Another hurdle is that while the most common perceptions about what diversity is are too weak to do the work set out for them, stronger notions of diversity are often defended on normative grounds that fail to connect to the values that are important to decision-makers in AI. However, there is a long history of research in feminist philosophy of science and a recent body of work in social epistemology that taken together provide the foundation for a notion of diversity that is both strong enough to do the work demanded of it, and can be defended on epistemic grounds that connect with the values that are important to decision-makers in AI. We clarify and defend that notion here by introducing emergent expertise as a network phenomenon wherein groups of workers with expertise of different types can gain knowledge not available to any individual alone, as long as they have ways of communicating across types of expertise. We illustrate the connected epistemic and ethical benefits of designing technology with diverse groups of workers using the examples of an infamous racist soap dispenser, and the millimeter wave scanners used in US airport security.

RCF-ST: RICHER CONVOLUTIONAL FEATURES NETWORK WITH STRUCTURAL TUNING FOR THE EDGE DETECTION ON NATURAL IMAGES

Context. The problem of automating of the edge detection on natural images in intelligent systems is considered. The subject of the research is the deep learning convolutional neural networks for edge detection on natural images.
 Objective. The objective of the research is to improve the edge detection performance of natural images by structural tuning the richer convolutional features network architecture.
 Method. In general, the edge detection performance is influenced by a neural network architecture. To automate the design of the network structure in the paper a structural tuning of a neural network is applied. Computational costs of a structural tuning are incomparably less compared with neural architecture search, but a higher qualification of the researcher is required, and the resulting solution will be suboptimal. In this research it is successively applied first a destructive approach and then a constructive approach to structural tuning of the based architecture of the RCF neural network. The constructive approach starts with a simple architecture network. Hidden layers, nodes, and connections are added to expand the network. The destructive approach starts with a complex architecture network. Hidden layers, nodes, and connections are then deleted to contract the network. The structural tuning of the richer convolutional features network includes: (1) reducing the number of convolutional layers; (2) reducing the number of convolutions in convolutional layers; (3) removing at each stage the sigmoid activation function with subsequent calculation of the loss function; (4) addition of the batch normalization layers after convolutional layers; (5) including the ReLU activation functions after the added batch normalization layers. The obtained neural network is named RCF-ST. The initial color images were scaled to the specified size and then inputted in the neural network. The advisability of each of the proposed stages of network structural tuning was reseached by estimating the edge detection performance using the confusion matrix elements and Figure of Merit. The advisability of a structural tuning of the neural network as a whole was estimated by comparing it with methods known from the literature using the Optimal Dataset Scale and Optimal Image Scale.
 Results. The proposed convolutional neural network has been implemented in software and researched for solving the problem of edge detection on natural images. The structural tuning technique may be used for informed design of the neural network architectures for other artificial intelligence problems.
 Conclusions. The obtained RCF-ST network allows to improve the performance of edge detection on natural images. RCF-ST network is characterized by a significantly fewer parameters compared to the RCF network, which makes it possible to reduce the resource consumption of the network. Besides, RCF-ST network ensures the enhancing of the robustness of edge detection on texture background.

Integrating prior knowledge to build transformer models

The big Artificial General Intelligence models inspire hot topics currently. The black box problems of Artificial Intelligence (AI) models still exist and need to be solved urgently, especially in the medical area. Therefore, transparent and reliable AI models with small data are also urgently necessary. To build a trustable AI model with small data, we proposed a prior knowledge-integrated transformer model. We first acquired prior knowledge using Shapley Additive exPlanations from various pre-trained machine learning models. Then, we used the prior knowledge to construct the transformer models and compared our proposed models with the Feature Tokenization Transformer model and other classification models. We tested our proposed model on three open datasets and one non-open public dataset in Japan to confirm the feasibility of our proposed methodology. Our results certified that knowledge-integrated transformer models perform better (1%) than general transformer models. Meanwhile, our proposed methodology identified that the self-attention of factors in our proposed transformer models is nearly the same, which needs to be explored in future work. Moreover, our research inspires future endeavors in exploring transparent small AI models.

Measuring Disentanglement: A Review of Metrics.

Learning to disentangle and represent factors of variation in data is an important problem in artificial intelligence. While many advances have been made to learn these representations, it is still unclear how to quantify disentanglement. While several metrics exist, little is known on their implicit assumptions, what they truly measure, and their limits. In consequence, it is difficult to interpret results when comparing different representations. In this work, we survey supervised disentanglement metrics and thoroughly analyze them. We propose a new taxonomy in which all metrics fall into one of the three families: intervention-based, predictor-based, and information-based. We conduct extensive experiments in which we isolate properties of disentangled representations, allowing stratified comparison along several axes. From our experiment results and analysis, we provide insights on relations between disentangled representation properties. Finally, we share guidelines on how to measure disentanglement.

Philosophical examination and thinking on the ethical problems of artificial intelligence

Philosophical examination and thinking on the ethical problems of artificial intelligence

Искусственный интеллект как часть общественной жизни: этико-правовые проблемы и пути их решения

The ethical and legal problems of artificial intelligence’s comprehensive integration into the public environment, its impact on the human life sphere are investigated. The intensity of the ongoing changes requires not only full-fledged legal regulation, but also a comprehensive study of both the positive aspects of innovations, as well as the foresight of possible threats and the formation of adequate protective mechanisms from the law point of view, from the standpoint of human moral values. It is a systematic approach and the possible negative consequences’ prevention of the artificial intelligence use that should shape the goal-setting of the development and implementation of such systems. Using scientific cognition methods, such as analysis, synthesis, system-analytical, formal-logical, system-legal, various approaches to understanding the definition of “artificial intelligence” have been studied, taking into account all the features of such systems, the necessary ways to solve the identified gaps in legal regulation related to determining the place of objects and artificial intelligence systems in the legal space. The social and practical significance of the development of robotics is touched upon, while attention is focused on the regulation of possible risks. It is concluded that it is the compliance of the functions performed by artificial intelligence with the ethical and moral canons of society that should be decisive in solving issues of responsibility. Separately, proposals are presented to determine civil liability both in cases of direct harm by robots with artificial intelligence, and in cases of damage or destruction of an artificial intelligence system by a third party to their owner.

Искусственный интеллект в политическом управлении: тенденции и перспективы

The article discusses the main directions of the use of artificial intelligence (AI) in the political life of society. It is noted that the level and quality of life of the population depends on the pace of AI implementation in various spheres of public life. The analysis of the conducted research in the format of an expert interview allows us to identify the key trends, advantages and challenges faced by the Russian political system in the implementation of AI. The results of the expert survey provide valuable practical recommendations and a basis for discussing possible directions for the development of AI in political management in Russia. The article notes the insufficient elaboration of the problems of artificial intelligence in the Russian scientific, in particular, political science discourse, which, in turn, is seen as promising in increasing research opportunities. The aim of the work is to identify the main trends in the use of artificial intelligence in political management and the prospects for its development in the political sphere of society. The consideration of the problem was based on the results of an initiative political and sociological study in the format of an expert interview (N=7). The following general scientific methods were used to analyze the results obtained: description, analysis, synthesis, classification. The study also used SWOT analysis to identify the strengths and weaknesses of using AI in the field of domestic state policy and electoral systems, as well as to show the opportunities and threats of its use. The analysis of the results of the expert survey showed that artificial intelligence is most in demand in such areas as analytics, decision-making, scenario forecasting, political communications, improving the efficiency of political institutions, especially the electoral system, monitoring voter sentiment, and in the field of security. Highlighting the shortcomings in the use of AI, experts noted the possible subjectivity and bias in decision-making, the problem of transparency and protection of personal data. As solutions to this problem, the development of appropriate regulations protecting the human right to confidentiality and privacy, "transparent" algorithms, and careful selection of data on the basis of which AI makes decisions were proposed. The prospect of further research lies in the development of ethical standards that can ensure transparency and security in the implementation of AI in political systems. An important conclusion of the article is the conclusion that there is an objective need to develop a regulatory framework and conduct regular inspections of AI systems to maintain trust and effective use of digital technologies in politics.

Ethics of Artificial Intelligence and Robotics: Key Issues and Modern Ways to Solve Them

Objective: modern achievements in the development and dissemination of digital technologies have attracted the attention of scholars and practitioners to the discussion of key ethical issues related to artificial intelligence and robotics. Hence, this study presents the most relevant of these issues, posing new challenges for legal scholars and practitioners to develop the regulation of artificial intelligence and robotics in terms of technology moralization.Methods: the research used practice- and risk-oriented approaches, complemented by multidisciplinary analysis of documents (European principles and codes of ethics) and studies, including those devoted to various problems of artificial intelligence and robotics.Results: the article identifies key ethical issues in the field of artificial intelligence and robotics. It is established that the key ethical issues involved can be solved if they are legally formalized and implemented at the international level. The algorithm proposed by the author, based on the analysis of the digital technologies application, will allow improving the moral actions of technologies in the process of their decision making.Scientific novelty: the article presents the latest ethical problems that concern scientists and practitioners in the field of artificial intelligence and robotics, and the methods of their solution by ethical and legal means aimed at moralizing technology and increasing its responsibility.Practical significance: all solutions presented in the article have practical significance and are ready for wide implementation at the international level. Their formalization in normative form and subsequent compliance will reduce the harm that artificial intelligence may cause in applied fields, including robotics using artificial intelligence. Regulatory, including legislative, decisions must therefore be taken as soon as possible to ensure that artificial intelligence and robotics become reliable tools for these systems to be used at work, at home, and in other areas such as shopping centers, stores, schools, universities, etc.

The Problematic Area of Philosophical Discourses on the Application of Artificial Intelligence Systems in Society

The study relevance lies in understanding strategic objectives' content concerning intelligent technologies’ application. The development and application of artificial intelligence in various branches of human activity carry the potential for global changes in society, which, in methodological terms, increases the relevance of considering these problems. The study of ethical problems of artificial intelligence in the concept of sustainable development of society is connected with the dynamic development of innovative artificial intelligence (AI) technologies, which are considered a process of becoming a new technosubject in the context of neuroscience development. The article emphasizes that the development of artificial intelligence acts as an innovative form of technology development when a new kind of human-machine technosubject is formed. The authors identify social problems of artificial intelligence application as the digitalization of life becomes a global anthropological challenge. Innovations of artificial intelligence are directly related to a number of ethical problems, as they contribute not only to the sustainable development of society but also cause dehumanization of relations between people and society. Based on the analysis of qualitative changes in scientific and technological research, using human-machine devices is becoming inevitable. Humankind must develop ethical guidelines for the application of AI systems.

Variatsionnye metody resheniya zadach, svyazannykh s iskusstvennym intellektom

We introduce the concepts of a learning algorithm, an objective function, a recognition system, a class of patterns, a training set, a reward algorithm, a finitely convergent algorithm, an adaptive control system, a control goal, control tactics, adaptation time, etc., related to the problem of artificial intelligence in the processes of learning and adaptation. The general problem of self-learning (unsupervised learning)—about the separation of sets—in terms of the classical calculus of variations is posed. The generality of the problem is due to the introduction of an additional time variable into the analysis. The problem is solved by determining extremal conditions under which the minimization of the overall average risk functional is achieved. Problems corresponding to nonfixed and fixed time intervals are considered. For these two cases, expressions are found for calculating variations in cost functionals. Necessary conditions are indicated for determining the extremal values of the self-learning process (separation of classes of a set of patterns) in time.

Экзистенциальные риски искусственного интеллекта в контексте аксиологии трансгуманизма (по материалам работ Н. Бострома)

The limitation on the development of artificial intelligence technologies which well-known developers of these technologies have recently been talking about, is due to fears of existential risks, one of which, according to the theory of the philosopher N. Bostrom, is Superintelligence as a result of the further development of artificial intelligence. Despite the fact that the problem of artificial intelligence has been widely studied in foreign philo-sophical literature, the issue of values underlying new technologies remains practically unstudied. The article substantiates that Bostrom's theory of risks of superintelligence is due to axiological reasons: artificial intelli-gence, according to the philosopher, can pose a threat to transhumanistic values. On the other hand, to prevent a catastrophe, it is these values that are supposed to be the basis for the development of artificial intelligence. The aim of the article is to analyze ethico-philosophical values of transhumanism and their role in the devel-opment of artificial intelligence technologies. It is noted that, in fact, current AI projects are developing precisely in the spirit of the philosophy of transhumanism. The scientific novelty of the article is due to the ethical and philosophical analysis of transhumanistic values, which N. Bostrom proposes to use in the context of the devel-opment of AI. As a result, the ethical meanings of the development of AI are assessed in the aspect of N. Bostrom’s concept of transhumanistic values.