Search

The eight papers included in this special issue represent a selection of extended contributions presented at the 17th International Conference on Soft Computing Models in Industrial and Environmental Applications, SOCO 2022 held in Salamanca, Spain, September 6th-8th, 2022, and organized by the BISITE group at University of Salamanca. SOCO 2022 international conference represents a collection or set of computational techniques in machine learning, computer science and some engineering disciplines which investigate, simulate, and analyse very complex issues and phenomena. This special issue is aimed at practitioners, researchers, and postgraduate students who are engaged in developing and applying advanced intelligent systems principles to solve real-world problems in the mentioned fields. [ABSTRACT FROM AUTHOR]

Achieving a widely accepted definition of human intelligence has been challenging, a situation mirrored by the diverse definitions of artificial intelligence in computer science. By critically examining published definitions, highlighting both consistencies and inconsistencies, this paper proposes a refined nomenclature that harmonizes conceptualizations across the two disciplines. Abstract and operational definitions for human and artificial intelligence are proposed that emphasize maximal capacity for completing novel goals successfully through respective perceptual-cognitive and computational processes. Additionally, support for considering intelligence, both human and artificial, as consistent with a multidimensional model of capabilities is provided. The implications of current practices in artificial intelligence training and testing are also described, as they can be expected to lead to artificial achievement or expertise rather than artificial intelligence. Paralleling psychometrics, 'AI metrics' is suggested as a needed computer science discipline that acknowledges the importance of test reliability and validity, as well as standardized measurement procedures in artificial system evaluations. Drawing parallels with human general intelligence, artificial general intelligence (AGI) is described as a reflection of the shared variance in artificial system performances. We conclude that current evidence more greatly supports the observation of artificial achievement and expertise over artificial intelligence. However, interdisciplinary collaborations, based on common understandings of the nature of intelligence, as well as sound measurement practices, could facilitate scientific innovations that help bridge the gap between artificial and human-like intelligence. • Proposes unified definitions for human and artificial intelligence. • Distinguishes between artificial achievement/expertise and artificial intelligence. • Advocates for AI metrics to ensure good quality AI system evaluations. • Describes artificial general intelligence (AGI) mirroring human general intelligence. • Evidence currently favours presence of artificial achievement over intelligence. [ABSTRACT FROM AUTHOR]

Combinatorial optimization problem (COP) is an NP-hard problem for which finding an optimal solution is difficult, especially as the problem size increases. The Traveling Salesman Problem (TSP), one of the COPs that can be formulated over a graph, is a well-researched area in operations research and computer science. Deep Reinforcement Learning (DRL) is now regarded as a promising approach for solving TSP and other NP hard problems. In this paper, we propose a novel Gated Deep Graph Attention Network (G-DGANet) which builds upon the existing Graph Neural Network (GNN) to solve TSP. The proposed G-DGANet uses gating mechanism between subsequent layers of the network to extract representations of nodes deeper in the network without loss in performance. G-DGANet also designs a novel aggregator to construct global graph embeddings from different embedding preferences. In addition, to effectively learn underlying structure of a graph, G-DGANet integrates node and edge information of the graph while updating node representations in the message passing mechanism. We used proximal policy optimization (PPO) to train G-DGANet on randomly generated instances. We conducted an experiment on randomly generated instances and on real-world road network data generated from digital maps to verify the performance of G-DGANet. The findings from experiments demonstrate that G-DGANet outperforms most traditional heuristics and existing DRL approaches, with high generalization abilities from random instance training to random instance testing and real-world road network instance testing. [ABSTRACT FROM AUTHOR]

The seven papers included in this special issue represent a selection of extended contributions presented at the 16th International Conference on Soft Computing Models in Industrial and Environmental Applications, SOCO 2021, held in Bilbao, Spain, September 22nd-24th, 2021, and organized by the BISITE group and the University of Deusto. The SOCO 2021 international conference represents a collection or set of computational techniques in machine learning, computer science and some engineering disciplines that investigate, simulate, and analyse very complex issues and phenomena. This special issue is aimed at practitioners, researchers, and postgraduate students who are engaged in developing and applying advanced intelligent systems principles to solve real-world problems in the mentioned fields. [ABSTRACT FROM AUTHOR]

Fair division of indivisible resources is a fundamental problem in many disciplines, including computer science, economy, operations research, etc. The existence of PMMS allocations of goods is a major open problem in fair division, even for additive valuations. At the state of the art, there is no identified setting where PMMS allocations are deemed impossible, and the known existing results regarding their existence are limited to highly constrained settings. In this paper, we provide an algorithm for computing a PMMS allocation for identical variant. We also use the algorithm to find the PMMS allocations for 3 agents in line graphs. Moreover, we show that a 4 5 -PMMS allocation can be computed in polynomial time when agents agree on the ordinal ranking of the goods. A quantitative gauge for assessing the impact on social welfare is known as the price of fairness , which quantifies the maximum possible reduction in social welfare due to the imposition of fairness constraints. Originally explored in the context of divisible goods, the concept of the price of fairness continues to be a pivotal metric for evaluating the effect of fairness considerations on overall social welfare. Consequently, we study the efficiency decrease under PMMS allocations and prove a tight upper bound on the price of PMMS allocations. • We present an algorithm to compute a PMMS allocation if agents' valuation functions are identical. • We analyze an existing algorithm terminates with a 4 5 -PMMS allocation in polynomial time when agents have identical ordinal ranking values of the goods. • We also examine the efficiency loss associated with PMMS allocations and establish their precise price bound. [ABSTRACT FROM AUTHOR]

In natural language processing (NLP), transformer based models have surpassed recurrent neural networks (RNN) as state of the art, being introduced specifically to address the limitations of RNNs originating from its sequential nature. As a similar sequence modeling problem, transformer methods can be readily adapted for deep learning time series prediction. This paper proposes a sparse transformer based approach for electricity load prediction. The layers of a transformer addresses the shortcomings of RNNs and CNNs by applying the attention mechanism on the entire time series, allowing any data point in the input to influence any location in the output of the layer. This allows transformers to incorporate information from the entire sequence in a single layer. Attention computations can also be parallelized. Thus, transformers can achieve faster speeds, or trade this speed for more layers and increased complexity. In experiments on public datasets, the sparse transformer attained comparable accuracy to an RNN-based SOTA method (Liu et al., 2022) while being up to 5× faster during inference. Moreover, the proposed model is general enough to forecast the load from individual households to city levels as shown in the extensive experiments conducted. [ABSTRACT FROM AUTHOR]

Artificial neural networks, inspired by the biological networks of the human brain, have become game-changing computing models in modern computer science. Inspired by their wide scope of applications, synthetic biology strives to create their biological counterparts, which we denote synthetic biological neural networks (SYNBIONNs). Their use in the fields of medicine, biosensors, biotechnology, and many more shows great potential and presents exciting possibilities. So far, many different synthetic biological networks have been successfully constructed, however, SYNBIONN implementations have been sparse. The latter are mostly based on neural networks pretrained in silico and being heavily dependent on extensive human input. In this paper, we review current implementations and models of SYNBIONNs. We briefly present the biological platforms that show potential for designing and constructing perceptrons and/or multilayer SYNBIONNs. We explore their future possibilities along with the challenges that must be overcome to successfully implement a scalable in vivo biological neural network capable of online learning. [ABSTRACT FROM AUTHOR]

Event sequence data is increasingly available. Many business operations are supported by information systems that record transactions, events, state changes, message exchanges, and similar elements. This observation also applies to various industries, including production, logistics, healthcare, financial services, and education. The variety of application areas explains that techniques for event sequence data analysis have been developed rather independently in different fields of computer science. Most prominent are contributions from information visualization and from process mining. So far, the contributions from these two fields have neither been compared nor have they been mapped to an integrated framework. Such intransparency is problematic since it bears the risk of opportunities of integration are missed and concepts established in one field are independently reinvented in the other one. In this paper, we develop the Event Sequence Visualization framework (ESeVis) that gives due credit to the traditions of both fields. Our mapping study provides an integrated perspective on both fields and identifies potential for synergies for future research. • We propose the ESeVis Framework to categorize visualizations of event sequence data. • We review the literature of event sequence data visualizations. • Instance representations of event sequences are mostly from information visualization. • Process mining focuses on generating formal models. • Potential for synergy of process mining and information visualization. [ABSTRACT FROM AUTHOR]

An eigenvalue λ of a graph G of order n is a main eigenvalue if its eigenspace is not orthogonal to the all-ones vector j n. In 1978, Cvetković proved that G has exactly one main eigenvalue if and only if G is regular, and posed the following long-standing problem: characterize the graphs with exactly k (2 ≤ k ≤ n) main eigenvalues. Graphs of order n with n , n − 1 main eigenvalues are called controllable, almost controllable, respectively. Cographs, threshold graphs are frequently studied in structural graph theory and computer science. In this paper, all almost controllable cographs, all almost controllable threshold graphs and all almost controllable graphs with second largest eigenvalue less than or equal to 5 − 1 2 are characterized. Furthermore, we give some results about cographs with exactly n − 2 main eigenvalues, and propose some additional problems for further study. [ABSTRACT FROM AUTHOR]