An artificial brain ATR's CAM-Brain Project aims to build/evolve an artificial brain with a million neural net modules inside a trillion cell Cellular Automata Machine

ATR's Evolutionary Systems Department aims to build (i.e. grow/evolve) an artificial brain by the year 2001. This artificial brain should initially contain thousands of interconnected artificial neural network modules, and be capable of controlling approximately 1000 “behaviors” in a “robot kitten”. The name given to this research project is “CAM-Brain”, because the neural networks (based on cellular automata) will be grown inside special hardware called Cellular Automata Machines (CAMs). Using a family of CAMs, each with its own processor to measure the performance quality or fitness of the evolved neural circuits, will allow the neural modules and their interconnections to be grown/evolved at electronic speeds. State of the art in CAM design is about 10 to the power 9 or 10 cells. Since a neural module of about 15 connected neurons can fit inside a cube of 100 cells on a side (1 million cells), a CAM which is specially adapted for CAM-Brain could contain thousands of interconnected modules, i.e. an artificial brain.     

Undecidable goals for completed acyclic programs

We prove for a well-known acyclic logic programP that it is undecidable whether or not a given goal is a logical consequence of the completion ofP. This complements recent decidability results for acyclic programs and bounded goals.     

Granularity analysis for exploiting adaptive parallelism of declarative programs on multiprocessors

Declarative Programming Languages (DPLs) apply a grocess model of Horn clauses such as PARLOG^[8] or a reduction model of λ-calculus such as SML^[7] and are,in principle,well suited to multiprocessor implementation.However,the performance of a paralled declarative program can be impaired by a mismatch between the parallelism available in an application and the parallelism available in the architecture.A particularly attractive solution is to automatically match the parallelism of the program to the parallelism of the target hardware as a compilation step.In this paper,we present an optimizing compilation technique called granularity analysis which identifies and removes excess parallelism that would degrade performance.The main steps are:an analysis of the flow of data to form an attributed call graph between function (or predicate) arguments;and an asymptotic estimation of granularity of a function (or predicate) to generate approximate grain size.Compiled procedure calls can be annotated with grain size and a task scheduler can make scheduling decisions with the classification scheme of grains to control parallelism at run-time.The resulting granularity analysis scheme is suitable for exploiting adaptive parallelism of declarative programming languages on multiprocessors.     

Implementing specifications by dynamic inheritance

An extended form of logic programming is presented which combines SLD-resolution with a dynamic form of inheritance between modules. The resulting inference system supports the notion of ‘implementation’ as a process of computing mappings between specifications and technologies—both formulated as sets of clausal-form theories—in order to satisfy requirements posed as queries. Besides detailing the operational features of the formalism, the paper also explores its semantics and its practical value in two case studies.     

The learning convergence of CMAC in cyclic learning

In this paper we discuss the learning convergence of the cerebellar model articulation controller (CMAC) in cyclic learning.We prove the following results.First,if the training samples are noiseless,the training algorithm converges if and only if the learning rate is chosen from (0,2).Second,when the training samples have noises,the learning algorithm will converge with a probability of one if the learning rate is dynamically decreased.Third,in the case with noises,with a small but fixed learning rate ε the mean square error of the weight sequences generated by the CMAC learning algorithm will be bounded by O(ε).Some simulation experiments are carried out to test these results.     

Distributed programming with Logic Tuple Spaces

From the point of view of distributed programming one of the most interesting communication mechanisms is associative tuple matching in a shared dataspace, as exemplified in the Linda coordination language. Linda has been used as a coordination layer to parallelize several sequential programming languages, such as C and Scheme. In this paper we study the combination of Linda with a logic language, whose result is the language Extended Shared Prolog (ESP). We show that ESP is based on a new programming model called PoliS, that extends Linda with Multiple Tuple Spaces. A class of applications for ESP is discussed, introducing the concept of “open multiple tuple spaces”. Finally, we show how the distributed implementation of ESP uses the network version of Linda’s tuple space.     

Domains of attraction in autoassociative memory networks

An autoassociative memory network is constructed by storing reference pattern vectors whose components consist of a small positive number ∈ and 1-∈. Although its connection weights can not be determined only by this storing condition, it is proved that the output function of the network becomes a contraction mapping in a region around each stored pattern if ∈ is sufficiently small. This implies that the region is a domain of attraction in the network. The shape of the region is clarified in our analysis. Domains of attraction larger than this region are also found. Any noisy pattern vector in such domains, which may have real valued components, can be recognized as one of the stored patterns. We propose a method for determining connection weights of the network, which uses the shape of the domains of attraction. The model obtained by this method has symmetric connection weights and is successfully applied to character pattern recognition.     

Iterative weighted least squares algorithms for neural networks classifiers

This paper discusses learning algorithms of layered neural networks from the standpoint of maximum likelihood estimation. At first we discuss learning algorithms for the most simple network with only one neuron. It is shown that Fisher information of the network, namely minus expected values of Hessian matrix, is given by a weighted covariance matrix of input vectors. A learning algorithm is presented on the basis of Fisher's scoring method which makes use of Fisher information instead of Hessian matrix in Newton's method. The algorithm can be interpreted as iterations of weighted least squares method. Then these results are extended to the layered network with one hidden layer. Fisher information for the layered network is given by a weighted covariance matrix of inputs of the network and outputs of hidden units. Since Newton's method for maximization problems has the difficulty when minus Hessian matrix is not positive definite, we propose a learning algorithm which makes use of Fisher information matrix, which is non-negative, instead of Hessian matrix. Moreover, to reduce the computation of full Fisher information matrix, we propose another algorithm which uses only block diagonal elements of Fisher information. The algorithm is reduced to an iterative weighted least squares algorithm in which each unit estimates its own weights by a weighted least squares method. It is experimentally shown that the proposed algorithms converge with fewer iterations than error back-propagation (BP) algorithm.     

On pseudo-BL-algebras and pseudo-hoops with normal maximal filters

We study the class of pseudo-BL-algebras whose every maximal filter is normal. We present an equational base for this class and we extend these results for the class of basic pseudo hoops with fixed strong unit. This is a continuation of the research from Botur et al. (Soft Comput 16:635–644, doi: 10.1007/s00500-011-0763-7, 2012).     

An extended Hamiltonian QR algorithm

An extended QR algorithm specifically tailored for Hamiltonian matrices is presented. The algorithm generalizes the customary Hamiltonian QR algorithm with additional freedom in choosing between various possible extended Hamiltonian Hessenberg forms. We introduced in Ferranti et al. (Calcolo, 2015. doi: 10.1007/s10092-016-0192-1) an algorithm to transform certain Hamiltonian matrices to such forms. Whereas the convergence of the classical QR algorithm is related to classical Krylov subspaces, convergence in the extended case links to extended Krylov subspaces, resulting in a greater flexibility, and possible enhanced convergence behavior. Details on the implementation, covering the bidirectional chasing and the bulge exchange based on rotations are presented. The numerical experiments reveal that the convergence depends on the selected extended forms and illustrate the validity of the approach.     

Comments on “A privacy preserving three-factor authentication protocol for e-health clouds”

The security provisioning of increasing healthcare data is of critical importance. The e-health clouds can be seen as a move towards an efficient management of such a big volume of healthcare data. Many schemes have been presented to bring more security and privacy along with efficiency, in the handling of data for booming e-health industry. Recently, in this connection, Jiang et al. (J Supercomput 1–24 doi: 10.1007/s11227-015-1610-x, 2016) presented a three-factor authentication protocol for e-health clouds. In this letter, we identify a serious flaw in the mutual authentication phase of the scheme, since an adversary may launch a denial-of-service attack (DoS) against the service providing server. Finally, we suggest a modification in the scheme to overcome the DoS attack.     

Towards an integrated formal method for verification of liveness properties in distributed systems: with application to population protocols

State-based formal methods [e.g. Event-B/RODIN (Abrial in Modeling in Event-B—system and software engineering. Cambridge University Press, Cambridge, 2010; Abrial et al. in Int J Softw Tools Technol Transf (STTT) 12(6):447–466, 2010)] for critical system development and verification are now well established, with track records including tool support and industrial applications. The focus of proof-based verification, in particular, is on safety properties. Liveness properties, which guarantee eventual, or converging computations of some requirements, are less well dealt with. Inductive reasoning about liveness is not explicitly supported. Liveness proofs are often complex and expensive, requiring high-skill levels on the part of the verification engineer. Fairness-based temporal logic approaches have been proposed to address this, e.g. TLA Lamport (ACM Trans Program Lang Syst 16(3):872–923, 1994) and that of Manna and Pnueli (Temporal verification of reactive systems—safety. Springer, New York, 1995). We contribute to this technology need by proposing a fairness-based method integrating temporal and first-order logic, proof and tools for modelling and verification of safety and liveness properties. The method is based on an integration of Event-B and TLA. Building on our previous work (Méry and Poppleton in Integrated formal methods, 10th international conference, IFM 2013, Turku, Finland, pp 208–222, 2013. doi: 10.1007/978-3-642-38613-8_15), we present the method via three example population protocols Angluin et al. (Distrib Comput 18(4):235–253, 2006). These were proposed as a theoretical framework for computability reasoning about Wireless Sensor Network and Mobile Ad-Hoc Network algorithms. Our examples present typical liveness and convergence requirements. We prove convergence results for the examples by integrated modelling and proof with Event-B/RODIN and TLA. We exploit existing proof rules, define and apply three new proof rules; soundness proofs are also provided. During the process we observe certain repeating patterns in the proofs. These are easily identified and reused because of the explicit nature of the reasoning.     

A commentary on “Hybrid search for minimal perturbation in Dynamic CSPs”

The Hybrid Search for Minimal Perturbation Problems algorithm in Dynamic CSP (HS_MPP) (Zivan, Constraints, 16(3), 228–249, 2011) ensures for a given dynamic problem and its solution to the previous CSP, to find the optimal solution to the newly generated CSP. This proposed method exploits the fact that its reported solution must satisfy two requirements. First, that it is a solution for T complete assignment for the derived CSP and second, that it is as close as possible to the solution of the former CSP. Unfortunately, the pseudo-code of the algorithm in Zivan (Constraints, 16(3), 228–249, 2011) is confusing and may lead to an implementation in which HS_MPP may not perform the expected outcomes of a given instance of Dynamic CSPs correctly. In this erratum, we demonstrate the possible undesired outcomes and give corrections in HS_MPP’s pseudo-code.     

Stochastic C-Stability and B-Consistency of Explicit and Implicit Milstein-Type Schemes

This paper focuses on two variants of the Milstein scheme, namely the split-step backward Milstein method and a newly proposed projected Milstein scheme, applied to stochastic differential equations which satisfy a global monotonicity condition. In particular, our assumptions include equations with super-linearly growing drift and diffusion coefficient functions and we show that both schemes are mean-square convergent of order 1. Our analysis of the error of convergence with respect to the mean-square norm relies on the notion of stochastic C-stability and B-consistency, which was set up and applied to Euler-type schemes in Beyn et al. (J Sci Comput 67(3):955–987, 2016. doi: 10.1007/s10915-015-0114-4). As a direct consequence we also obtain strong order 1 convergence results for the split-step backward Euler method and the projected Euler–Maruyama scheme in the case of stochastic differential equations with additive noise. Our theoretical results are illustrated in a series of numerical experiments.     

A hybrid deep learning neural approach for emotion recognition from facial expressions for socially assistive robots

We have recently seen significant advancements in the development of robotic machines that are designed to assist people with their daily lives. Socially assistive robots are now able to perform a number of tasks autonomously and without human supervision. However, if these robots are to be accepted by human users, there is a need to focus on the form of human–robot interaction that is seen as acceptable by such users. In this paper, we extend our previous work, originally presented in Ruiz-Garcia et al. (in: Engineering applications of neural networks: 17th international conference, EANN 2016, Aberdeen, UK, September 2–5, 2016, proceedings, pp 79–93, 2016. https://doi.org/10.1007/978-3-319-44188-7_6), to provide emotion recognition from human facial expressions for application on a real-time robot. We expand on previous work by presenting a new hybrid deep learning emotion recognition model and preliminary results using this model on real-time emotion recognition performed by our humanoid robot. The hybrid emotion recognition model combines a Deep Convolutional Neural Network (CNN) for self-learnt feature extraction and a Support Vector Machine (SVM) for emotion classification. Compared to more complex approaches that use more layers in the convolutional model, this hybrid deep learning model produces state-of-the-art classification rate of \(96.26\%\), when tested on the Karolinska Directed Emotional Faces dataset (Lundqvist et al. in The Karolinska Directed Emotional Faces—KDEF, 1998), and offers similar performance on unseen data when tested on the Extended Cohn–Kanade dataset (Lucey et al. in: Proceedings of the third international workshop on CVPR for human communicative behaviour analysis (CVPR4HB 2010), San Francisco, USA, pp 94–101, 2010). This architecture also takes advantage of batch normalisation (Ioffe and Szegedy in Batch normalization: accelerating deep network training by reducing internal covariate shift. http://arxiv.org/abs/1502.03167, 2015) for fast learning from a smaller number of training samples. A comparison between Gabor filters and CNN for feature extraction, and between SVM and multilayer perceptron for classification is also provided.

     

Image Colorization Based on a Generalization of the Low Dimensional Manifold Model

In this paper, we introduce a novel model that restores a color image from a grayscale image with color values given in small regions. The model is based on the idea of the generalization of the low dimensional manifold model (Shi et al. in J Sci Comput, 2017.  https://doi.org/10.1007/s10915-017-0549-x) and the YCbCr color space. It involves two prior terms, a weighted nonlocal Laplacian (WNLL) and a weighted total variation (WTV). The WNLL allows regions without color information to be interpolated smoothly from given sparse color data, while the WTV assists to inhibit the diffusion of color values across edges. To cope with various types of sampled data, we introduce an updating rule for the weight function in the WNLL. Furthermore, we present an efficient iterative algorithm for solving the proposed model. Lastly, numerical experiments validate the superior performance of the proposed model over that of the other state-of-the-art models.     

Mean-square stability and error analysis of implicit time-stepping schemes for linear parabolic SPDEs with multiplicative Wiener noise in the first derivative

In this article, we extend a Milstein finite difference scheme introduced in 8 Giles, M. B. and Reisinger, C. 2012. Stochastic finite differences and multilevel Monte Carlo for a class of SPDEs in finance. SIAM Financ. Math., 3(1): 572–592. (doi:10.1137/110841916)[Crossref] , [Google Scholar] for a certain linear stochastic partial differential equation (SPDE) to semi-implicit and fully implicit time-stepping as introduced by Szpruch 32 Szpruch, L. 2010. Numerical approximations of nonlinear stochastic systems PhD Thesis, University of Strathclyde [Google Scholar] for stochastic differential equations (SDEs). We combine standard finite difference Fourier analysis for partial differential equations with the linear stability analysis in 3 Buckwar, E. and Sickenberger, T. 2011. A comparative linear mean-square stability analysis of Maruyama- and Milstein-type methods. Math. Comput. Simulation, 81: 1110–1127. (doi:10.1016/j.matcom.2010.09.015)[Crossref], [Web of Science ®] , [Google Scholar] for SDEs to analyse the stability and accuracy. The results show that Crank–Nicolson time-stepping for the principal part of the drift with a partially implicit but negatively weighted double Itô integral gives unconditional stability over all parameter values and converges with the expected order in the mean-square sense. This opens up the possibility of local mesh refinement in the spatial domain, and we show experimentally that this can be beneficial in the presence of reduced regularity at boundaries.     

Two-factor authentication for the Bitcoin protocol

We show how to realize two-factor authentication for a Bitcoin wallet. To do so, we explain how to employ an ECDSA adaption of the two-party signature protocol by MacKenzie and Reiter (Int J Inf Secur 2(3–4):218–239, 2004. doi: 10.1007/s10207-004-0041-0) in the context of Bitcoin and present a prototypic implementation of a Bitcoin wallet that offers both: two-factor authentication and verification over a separate channel. Since we use a smart phone as the second authentication factor, our solution can be used with hardware already available to most users and the user experience is quite similar to the existing online banking authentication methods.     

Evolutionary Frequency and Forecasting Accuracy: Simulations Based on an Agent-Based Artificial Stock Market

Learning how to forecast is always important for traders, and divergent learning frequencies prevail among traders. The influence of the evolutionary frequency on learning performance has occasioned many studies of agent-based computational finance (e.g., Lettau in J Econ Dyn Control 21:1117–1147, 1997. doi: 10.1016/S0165-1889(97)00046-8; Szpiro in Complexity 2(4):31–39, 1997. doi: 10.1002/(SICI)1099-0526(199703/04)2:4<31::AID-CPLX8>3.0.CO;2-3; Cacho and Simmons in Aust J Agric Resour Econ 43(3):305–322, 1999. doi: 10.1111/1467-8489.00081). Although these studies all suggest that evolving less frequently and, hence, experiencing more realizations help learning, this implication may result from their common stationary assumption. Therefore, we first attempt to approach this issue in a ‘dynamically’ evolving market in which agents learn to forecast endogenously generated asset prices. Moreover, in these studies’ market settings, evolving less frequently also meant having a longer time horizon. However, it is not true in many market settings that are even closer to the real financial markets. The clarification that the evolutionary frequency and the time horizon are two separate notions leaves the effect of the evolutionary frequency on learning even more elusive and worthy of exploration independently. We find that the influence of a trader’s evolutionary frequency on his forecasting accuracy depends on all market participants and the resulting price dynamics. In addition, prior studies also commonly assume that traders have identical preferences, which is too strong an assumption to apply to a real market. Considering the heterogeneity of preferences, we find that converging to the rational expectations equilibrium is hardly possible, and we even suggest that agents in a slow-learning market learn frequently. We also apply a series of econometric tests to explain the simulation results.     

Linearity and iterator types for Gödel’s System

System Open image in new window is a linear λ-calculus with numbers and an iterator, which, although imposing linearity restrictions on terms, has all the computational power of Gödel’s System  Open image in new window . System Open image in new window owes its power to two features: the use of a closed reduction strategy (which permits the construction of an iterator on an open function, but only iterates the function after it becomes closed), and the use of a liberal typing rule for iterators based on iterative types. In this paper, we study these new types, and show how they relate to intersection types. We also give a sound and complete type reconstruction algorithm for System  Open image in new window .