Search Results (7)

The input layer, hidden layer, and output layer are three models of the neural processors that make up feedforward neural networks (FNNs). Evolutionary algorithms have been extensively employed in training FNNs, which can correctly actualize any finite training sample set. In this paper, an enhanced marine predators algorithm (MPA) based on the ranking-based mutation operator (EMPA) was presented to train FNNs, and the objective was to attain the minimum classification, prediction, and approximation errors by modifying the connection weight and deviation value. The ranking-based mutation operator not only determines the best search agent and elevates the exploitation ability, but it also delays premature convergence and accelerates the optimization process. The EMPA integrates exploration and exploitation to mitigate search stagnation, and it has sufficient stability and flexibility to acquire the finest solution. To assess the significance and stability of the EMPA, a series of experiments on seventeen distinct datasets from the machine learning repository of the University of California Irvine (UCI) were utilized. The experimental results demonstrated that the EMPA has a quicker convergence speed, greater calculation accuracy, higher classification rate, strong stability and robustness, which is productive and reliable for training FNNs. Full article

How can single cells without nervous systems perform complex behaviours such as habituation, associative learning and decision making, which are considered the hallmark of animals with a brain? Are there molecular systems that underlie cognitive properties equivalent to those of the brain? This review follows the development of the idea of molecular brains from Darwin’s “root brain hypothesis”, through bacterial chemotaxis, to the recent discovery of neuron-like r-protein networks in the ribosome. By combining a structural biology view with a Bayesian brain approach, this review explores the evolutionary labyrinth of information processing systems across scales. Ribosomal protein networks open a window into what were probably the earliest signalling systems to emerge before the radiation of the three kingdoms. While ribosomal networks are characterised by long-lasting interactions between their protein nodes, cell signalling networks are essentially based on transient interactions. As a corollary, while signals propagated in persistent networks may be ephemeral, networks whose interactions are transient constrain signals diffusing into the cytoplasm to be durable in time, such as post-translational modifications of proteins or second messenger synthesis. The duration and nature of the signals, in turn, implies different mechanisms for the integration of multiple signals and decision making. Evolution then reinvented networks with persistent interactions with the development of nervous systems in metazoans. Ribosomal protein networks and simple nervous systems display architectural and functional analogies whose comparison could suggest scale invariance in information processing. At the molecular level, the significant complexification of eukaryotic ribosomal protein networks is associated with a burst in the acquisition of new conserved aromatic amino acids. Knowing that aromatic residues play a critical role in allosteric receptors and channels, this observation suggests a general role of π systems and their interactions with charged amino acids in multiple signal integration and information processing. We think that these findings may provide the molecular basis for designing future computers with organic processors. Full article

In this paper, we propose direct simulations between a given network of evolutionary processors with an arbitrary topology of the underlying graph and a network of evolutionary processors with underlying graphs—that is, a complete graph, a star graph and a grid graph, respectively. All of these simulations are time complexity preserving—namely, each computational step in the given network is simulated by a constant number of computational steps in the constructed network. These results might be used to efficiently convert a solution of a problem based on networks of evolutionary processors provided that the underlying graph of the solution is not desired. Full article

This paper compares the latency, accuracy, training time and hardware costs of neural networks compressed with our new multi-objective evolutionary algorithm called NEMOKD, and with quantisation. We evaluate NEMOKD on Intel’s Movidius Myriad X VPU processor, and quantisation on Xilinx’s programmable Z7020 FPGA hardware. Evolving models with NEMOKD increases inference accuracy by up to 82% at the cost of 38% increased latency, with throughput performance of 100–590 image frames-per-second (FPS). Quantisation identifies a sweet spot of 3 bit precision in the trade-off between latency, hardware requirements, training time and accuracy. Parallelising FPGA implementations of 2 and 3 bit quantised neural networks increases throughput from 6 k FPS to 373 k FPS, a 62$\times$ speedup. Full article

Networks of picture processors is a massively distributed and parallel computational model inspired by the evolutionary cellular processes, which offers efficient solutions for NP-complete problems. This bio-inspired model computes two-dimensional strings (pictures) using simple rewriting rules (evolutionary operations). The functioning of this model mimics a community of cells (pictures) that are evolving according to these bio-operations via a selection process that filters valid surviving cells. In this paper, we propose an extension of this model that empowers it with a flexible method that selects the processed pictures based on a quantitative evaluation of its content. In order to show the versatility of this extension, we introduce a solver for a cryptographic proof-of-work based on the hardness of finding a solution to a set of random quadratic equations over the finite field ${\mathbb{F}}_2$. This problem is demonstrated to be NP-hard, even with quadratic polynomials over the field ${\mathbb{F}}_2$, when the number of equations and the number of variables are of roughly the same size. The proposed solution runs in $O\left(n^2\right)$ computational steps for any size $(n,m)$ of the input pictures. In this context, this paper opens up a wide field of research that looks for theoretical and practical solutions of cryptographic problems via software/hardware implementations based on bio-inspired computational models. Full article

This paper describes one grid-based genetic algorithm approach to solve the vehicle routing problem with time windows in one experimental cluster MiniGrid. Clusters used in this approach are located in two Mexican cities (Cuernavaca and Jiutepec, Morelos) securely communicating with each other since they are configured as one virtual private network, and its use as a single set of processors instead of isolated groups allows one to increase the computing power to solve complex tasks. The genetic algorithm splits the population of candidate solutions in several segments, which are simultaneously mutated in each process generated by the MiniGrid. These mutated segments are used to build a new population combining the results produced by each process. In this paper, the MiniGrid configuration scheme is described, and both the communication latency and the speedup behavior are discussed. Experimental results show one information exchange reduction through the MiniGrid clusters as well as an improved behavior of the evolutionary algorithm. A statistical analysis of these results suggests that our approach is better as a combinatorial optimization procedure as compared with other methods. Full article

In the optimization of deep neural networks (DNNs) via evolutionary algorithms (EAs) and the implementation of the training necessary for the creation of the objective function, there is often a trade-off between efficiency and flexibility. Pure software solutions implemented on general-purpose processors tend to be slow because they do not take advantage of the inherent parallelism of these devices, whereas hardware realizations based on heterogeneous platforms (combining central processing units (CPUs), graphics processing units (GPUs) and/or field-programmable gate arrays (FPGAs)) are designed based on different solutions using methodologies supported by different languages and using very different implementation criteria. This paper first presents a study that demonstrates the need for a heterogeneous (CPU-GPU-FPGA) platform to accelerate the optimization of artificial neural networks (ANNs) using genetic algorithms. Second, the paper presents implementations of the calculations related to the individuals evaluated in such an algorithm on different (CPU- and FPGA-based) platforms, but with the same source files written in OpenCL. The implementation of individuals on remote, low-cost FPGA systems on a chip (SoCs) is found to enable the achievement of good efficiency in terms of performance per watt. Full article