Search Results (35)

Parallel Communicating Finite Automata (PCFA) extend classical finite automata by enabling multiple automata to operate in parallel and communicate upon request, capturing essential aspects of parallel and distributed computation. This model is relevant for studying complex systems such as computer networks and multi-agent environments. In this paper, we explore two key aspects of PCFA: their undecidability and their descriptional complexity. We first show that deterministic PCFA of degree 2 ($DPCFA\left(2\right)$) can accept a set of valid computations of a deterministic Turing machine, leading to the undecidability of restricted versions of emptiness and universality problems. Additionally, we employ the concept of productiveness (a stronger form of non-recursive enumerability) to demonstrate that these problems are not only undecidable but also unprovable. Second, we investigate the descriptional complexity of PCFA and establish non-recursive trade-offs between different PCFA models and many classes of language descriptors, such as DFAs and subclasses of regular expressions, offering new insights into their computational and structural properties. Full article

Machine learning has become indispensable across various domains, yet understanding its theoretical underpinnings remains challenging for many practitioners and researchers. Despite the availability of numerous resources, there is a need for a cohesive tutorial that integrates foundational principles with state-of-the-art theories. This paper addresses the fundamental concepts and theories of machine learning, with an emphasis on neural networks, serving as both a foundational exploration and a tutorial. It begins by introducing essential concepts in machine learning, including various learning and inference methods, followed by criterion functions, robust learning, discussions on learning and generalization, model selection, bias–variance trade-off, and the role of neural networks as universal approximators. Subsequently, the paper delves into computational learning theory, with probably approximately correct (PAC) learning theory forming its cornerstone. Key concepts such as the VC-dimension, Rademacher complexity, and empirical risk minimization principle are introduced as tools for establishing generalization error bounds in trained models. The fundamental theorem of learning theory establishes the relationship between PAC learnability, Vapnik–Chervonenkis (VC)-dimension, and the empirical risk minimization principle. Additionally, the paper discusses the no-free-lunch theorem, another pivotal result in computational learning theory. By laying a rigorous theoretical foundation, this paper provides a comprehensive tutorial for understanding the principles underpinning machine learning. Full article

The mathematical representation of the universe consists of sequences of symbols, rules and operators containing Gödel’s undecidable propositions: information and its manipulation, also with Turing Machines. Classical information theory and mathematics, ideally independent from the medium used, can be interpreted realistically and objectively from their correspondence with quantum information, which is physical. Each representation of the universe and its evolution are, in any case, physical subsets of the universe, structured sets of observers and their complements in the universe made with spacetime events generated by local quantum measurements. Their description becomes a semantically closed structure without a global object-environment loss of decoherence as a von Neumann’s universal constructor with a semantical abstract whose structure cannot be decided deterministically a priori from an internal observer. In a semantically closed structure, the realization of a specific event that writes the semantical abstract of the constructor is a problem of finding “which way” for the evolution of the universe as a choice of the constructor’s state in a metastructure, like the many-world Everett scenario, from a specific result of any quantum measurement, corresponding to a Gödel undecidable proposition for an internal observer. Full article

The integration of deep learning into radiology has the potential to enhance diagnostic processes, yet its acceptance in clinical practice remains limited due to various challenges. This study aimed to develop and evaluate a fine-tuned large language model (LLM), based on Llama 3-8B, to automate the generation of accurate and concise conclusions in magnetic resonance imaging (MRI) and computed tomography (CT) radiology reports, thereby assisting radiologists and improving reporting efficiency. A dataset comprising 15,000 radiology reports was collected from the University of Medicine and Pharmacy of Craiova’s Imaging Center, covering a diverse range of MRI and CT examinations made by four experienced radiologists. The Llama 3-8B model was fine-tuned using transfer-learning techniques, incorporating parameter quantization to 4-bit precision and low-rank adaptation (LoRA) with a rank of 16 to optimize computational efficiency on consumer-grade GPUs. The model was trained over five epochs using an NVIDIA RTX 3090 GPU, with intermediary checkpoints saved for monitoring. Performance was evaluated quantitatively using Bidirectional Encoder Representations from Transformers Score (BERTScore), Recall-Oriented Understudy for Gisting Evaluation (ROUGE), Bilingual Evaluation Understudy (BLEU), and Metric for Evaluation of Translation with Explicit Ordering (METEOR) metrics on a held-out test set. Additionally, a qualitative assessment was conducted, involving 13 independent radiologists who participated in a Turing-like test and provided ratings for the AI-generated conclusions. The fine-tuned model demonstrated strong quantitative performance, achieving a BERTScore F1 of 0.8054, a ROUGE-1 F1 of 0.4998, a ROUGE-L F1 of 0.4628, and a METEOR score of 0.4282. In the human evaluation, the artificial intelligence (AI)-generated conclusions were preferred over human-written ones in approximately 21.8% of cases, indicating that the model’s outputs were competitive with those of experienced radiologists. The average rating of the AI-generated conclusions was 3.65 out of 5, reflecting a generally favorable assessment. Notably, the model maintained its consistency across various types of reports and demonstrated the ability to generalize to unseen data. The fine-tuned Llama 3-8B model effectively generates accurate and coherent conclusions for MRI and CT radiology reports. By automating the conclusion-writing process, this approach can assist radiologists in reducing their workload and enhancing report consistency, potentially addressing some barriers to the adoption of deep learning in clinical practice. The positive evaluations from independent radiologists underscore the model’s potential utility. While the model demonstrated strong performance, limitations such as dataset bias, limited sample diversity, a lack of clinical judgment, and the need for large computational resources require further refinement and real-world validation. Future work should explore the integration of such models into clinical workflows, address ethical and legal considerations, and extend this approach to generate complete radiology reports. Full article

Genetic programming (GP) has a long-standing tradition in the evolution of computer programs, predominantly utilizing tree and linear paradigms, each with distinct advantages and limitations. Despite the rapid growth of the GP field, there have been disproportionately few attempts to evolve ’real’ Turing-like imperative programs (as contrasted with functional programming) from the ground up. Existing research focuses mainly on specific special cases where the structure of the solution is partly known. This paper explores the potential of integrating tree and linear GP paradigms to develop an encoding scheme that universally supports genetic operators without constraints and consistently generates syntactically correct Python programs from scratch. By blending the symmetrical structure of tree-based representations with the inherent asymmetry of linear sequences, we created a versatile environment for program evolution. Our approach was rigorously tested on 35 problems characterized by varying Halstead complexity metrics, to delineate the approach’s boundaries. While expected brute-force program solutions were observed, our method yielded more sophisticated strategies, such as optimizing a program by restricting the division trials to the values up to the square root of the number when counting its proper divisors. Despite the recent groundbreaking advancements in large language models, we assert that the GP field warrants continued research. GP embodies a fundamentally different computational paradigm, crucial for advancing our understanding of natural evolutionary processes. Full article

The dichotomy in power consumption between digital and biological information processing systems is an intriguing open question related at its core with the necessity for a more thorough understanding of the thermodynamics of the logic of computing. To contribute in this regard, we put forward a model that implements the Boltzmann machine (BM) approach to computation through an electric substrate under thermal fluctuations and dissipation. The resulting network has precisely defined statistical properties, which are consistent with the data that are accessible to the BM. It is shown that by the proposed model, it is possible to design neural-inspired logic gates capable of universal Turing computation under similar thermal conditions to those found in biological neural networks and with information processing and storage electric potentials at comparable scales. Full article

Spiking neural P systems (SNP systems), as computational models abstracted by the biological nervous system, have been a major research topic in biological computing. In conventional SNP systems, the rules in a neuron remain unchanged during the computation. In the biological nervous system, however, the biochemical reactions in a neuron are also influenced by factors such as the substances contained in it. Based on this motivation, this paper proposes SNP systems with rules dynamic generation and removal (RDGRSNP systems). In RDGRSNP systems, the application of rules leads to changes of the substances in neurons, which leads to changes of the rules in neurons. The Turing universality of RDGRSNP systems is demonstrated as a number-generating device and a number-accepting device, respectively. Finally, a small universal RDGRSNP system for function computation using 68 neurons is given. It is demonstrated that the variant we proposed requires fewer neurons by comparing it with five variants of SNP systems. Full article

Can free agency be compatible with determinism? Compatibilists argue that the answer is yes, and it has been suggested that the computer science principle of “computational irreducibility” sheds light on this compatibility. It implies that there cannot, in general, be shortcuts to predict the behavior of agents, explaining why deterministic agents often appear to act freely. In this paper, we introduce a variant of computational irreducibility that intends to capture more accurately aspects of actual (as opposed to apparent) free agency, including computational sourcehood, i.e., the phenomenon that the successful prediction of a process’ behavior must typically involve an almost-exact representation of the relevant features of that process, regardless of the time it takes to arrive at the prediction. We argue that this can be understood as saying that the process itself is the source of its actions, and we conjecture that many computational processes have this property. The main contribution of this paper is technical, in that we analyze whether and how a sensible formal definition of computational sourcehood is possible. While we do not answer the question completely, we show how it is related to finding a particular simulation preorder on Turing machines, we uncover concrete stumbling blocks towards constructing such a definition, and demonstrate that structure-preserving (as opposed to merely simple or efficient) functions between levels of simulation play a crucial role. Full article

In this work, a color image encryption and decryption algorithm for digital images is presented. It is based on the modular discrete derivative (MDD), a novel technique to encrypt images and efficiently hide visual information. In addition, Langton’s ant, which is a two-dimensional universal Turing machine with a high key space, is used. Moreover, a deterministic noise technique that adds security to the MDD is utilized. The proposed hybrid scheme exploits the advantages of MDD and Langton’s ant, generating a very secure and reliable encryption algorithm. In this proposal, if the key is known, the original image is recovered without loss. The method has demonstrated high performance through various tests, including statistical analysis (histograms and correlation distributions), entropy, texture analysis, encryption quality, key space assessment, key sensitivity analysis, and robustness to differential attack. The proposed method highlights obtaining chi-square values between $233.951$ and $281.687$, entropy values between $7.9999225223$ and $7.9999355791$, PSNR values (in the original and encrypted images) between $8.134$ and $9.957$, the number of pixel change rate (NPCR) values between $99.60851796\%$ and $99.61054611\%$, unified average changing intensity (UACI) values between $33.44672377\%$ and $33.47430379\%$, and a vast range of possible keys $>5.8459\times {10}^{72}$. On the other hand, an analysis of the sensitivity of the key shows that slight changes to the key do not generate any additional information to decrypt the image. In addition, the proposed method shows a competitive performance against recent works found in the literature. Full article

This article provides a detailed analysis of the transfer of a key cluster of ideas from mathematical logic to computing. We demonstrate the impact of certain of Turing’s logico-philosophical concepts from the mid-1930s on the emergence of the modern electronic computer—and so, in consequence, Turing’s impact on the direction of modern philosophy, via the computational turn. We explain why both Turing and von Neumann saw the problem of developing the electronic computer as a problem in logic, and we describe their joint journey from logic to electronic computation. While much has been written about Turing’s and von Neumann’s individual contributions to the development of the computer, this article investigates less well-known terrain: their interactions and mutual influences. Along the way we argue against ‘logic skeptics’ and ‘Turing skeptics’, who claim that neither logic nor Turing played any significant role in the creation of the modern computer. Full article

Cell-like P systems with channel states and symport/antiport rules (CCS P systems) are a type of nondeterministic parallel biological computing model, where there exists a channel between adjacent regions and there is a state on each channel to control the execution of symport/antiport rules. In this work, a synchronization rule is introduced into CCS P systems, a variant of CCS P systems called CCS P systems with synchronization rule (CCSs P systems) is proposed. The universality of CCSs P systems with only uniport (symport or antiport) rules is investigated. By simulating the register machine, we proved that CCSs P systems have the ability to simulate any Turing machine in the following three cases: having two membranes, two channel states and using symport rules of length at most 2; having one membrane, three channel states and using symport rules of length at most 2; and having one membrane, two channel states and using antiport rules of length at most 3. Full article

P systems with active membranes, as a sort of basic P system, include in communication rules and out communication rules, where communication rules are controlled by polarizations. However, the communication of objects among living cells may be controlled by several factors, such as proteins, polarizations, etc. Based on this biological fact, in this article, a new class of P systems, named P systems with proteins on active membranes (known as PAM P systems) is considered, where the movement of objects is controlled by both proteins and polarizations. The computational theory of PAM P systems is discussed. More specifically, we show that PAM P systems achieve Turing universality when the systems use two membranes, one protein and one polarization. Moreover, the PAM P systems, with the help of membrane division rules, make the SAT problem solvable. These results indicate that PAM P systems are also a sort of powerful system. Full article

As one of the key topics in the development of neighborhood rough set, attribute reduction has attracted extensive attentions because of its practicability and interpretability for dimension reduction or feature selection. Although the random sampling strategy has been introduced in attribute reduction to avoid overfitting, uncontrollable sampling may still affect the efficiency of search reduct. By utilizing inherent characteristics of each label, Multi-label learning with Label specIfic FeaTures (Lift) algorithm can improve the performance of mathematical modeling. Therefore, here, it is attempted to use Lift algorithm to guide the sampling for reduce the uncontrollability of sampling. In this paper, an attribute reduction algorithm based on Lift and random sampling called ARLRS is proposed, which aims to improve the efficiency of searching reduct. Firstly, Lift algorithm is used to choose the samples from the dataset as the members of the first group, then the reduct of the first group is calculated. Secondly, random sampling strategy is used to divide the rest of samples into groups which have symmetry structure. Finally, the reducts are calculated group-by-group, which is guided by the maintenance of the reducts’ classification performance. Comparing with other 5 attribute reduction strategies based on rough set theory over 17 University of California Irvine (UCI) datasets, experimental results show that: (1) ARLRS algorithm can significantly reduce the time consumption of searching reduct; (2) the reduct derived from ARLRS algorithm can provide satisfying performance in classification tasks. Full article

Eco-cognitive computationalism explores computing in context, adhering to some of the key ideas presented by modern cognitive science perspectives on embodied, situated, and distributed cognition. First of all, when physical computation is seen from the perspective of the ecology of cognition it is possible to clearly understand the role Turing assigned to the process of “education” of the machine, paralleling it to the education of human brains, in the invention of the Logical Universal Machine. It is this Turing’s emphasis on education that furnishes the justification of the conceptualization of machines as “domesticated ignorant entities”, that is proposed in this article. I will show that conceptualizing machines as dynamically active in distributed physical entities of various kinds suitably transformed so that data can be encoded and decoded to obtain appropriate results sheds further light on my eco-cognitive perspective. Furthermore, it is within this intellectual framework that I will usefully analyze the recent attention in computer science devoted to the importance of the simplification of cognitive and motor tasks caused in organic entities thanks to morphological features: ignorant bodies can be computationally domesticated to make an intertwined computation simpler, relying on the “simplexity” of animal embodied cognition, which represents one of the main qualities of organic agents. Finally, eco-cognitive computationalism allows us to clearly acknowledge that the concept of computation evolves over time as a result of historical and contextual factors, and we can construct an epistemological view that depicts the “emergence” of new types of computations that exploit new substrates. This new viewpoint demonstrates how the computational domestication of ignorant entities might result in the emergence of novel unconventional cognitive embodiments. Full article

Spiking neural P systems (SN P systems for short) realize the high abstraction and simulation of the working mechanism of the human brain, and adopts spikes for information encoding and processing, which are regarded as one of the third-generation neural network models. In the nervous system, the conduction of excitation depends on the presence of membrane potential (also known as the transmembrane potential difference), and the conduction of excitation on neurons is the conduction of action potentials. On the basis of the SN P systems with polarizations, in which the neuron-associated polarization is the trigger condition of the rule, the concept of neuronal membrane potential is introduced into systems. The obtained variant of the SN P system features charge accumulation and computation within neurons in quantity, as well as transmission between neurons. In addition, there are inhibitory synapses between neurons that inhibit excitatory transmission, and as such, synapses cause postsynaptic neurons to generate inhibitory postsynaptic potentials. Therefore, to make the model better fit the biological facts, inhibitory rules and anti-spikes are also adopted to obtain the spiking neural P systems with membrane potentials, inhibitory rules, and anti-spikes (referred to as the MPAIRSN P systems). The Turing universality of the MPAIRSN P systems as number generating and accepting devices is demonstrated. On the basis of the above working mechanism of the system, a small universal MPAIRSN P system with 95 neurons for computing functions is designed. The comparisons with other SN P models conclude that fewer neurons are required by the MPAIRSN P systems to realize universality. Full article