Comparative Experimental Studies on Spatial Memory and Learning in Rats and Robots

The study of behavioral and neurophysiological mechanisms involved in rat spatial cognition provides a basis for the development of computational models and robotic experimentation of goal-oriented learning tasks. These models and robotics architectures offer neurobiologists and neuroethologists alternative platforms to study, analyze and predict spatial cognition based behaviors. In this paper we present a comparative analysis of spatial cognition in rats and robots by contrasting similar goal-oriented tasks in a cyclical maze, where studies in rat spatial cognition are used to develop computational system-level models of hippocampus and striatum integrating kinesthetic and visual information to produce a cognitive map of the environment and drive robot experimentation. During training, Hebbian learning and reinforcement learning, in the form of Actor-Critic architecture, enable robots to learn the optimal route leading to a goal from a designated fixed location in the maze. During testing, robots exploit maximum expectations of reward stored within the previously acquired cognitive map to reach the goal from different starting positions. A detailed discussion of comparative experiments in rats and robots is presented contrasting learning latency while characterizing behavioral procedures during navigation such as errors associated with the selection of a non-optimal route, body rotations, normalized length of the traveled path, and hesitations. Additionally, we present results from evaluating neural activity in rats through detection of the immediate early gene Arc to verify the engagement of hippocampus and striatum in information processing while solving the cyclical maze task, such as robots use our corresponding models of those neural structures.     

Hidden Markov models for the stimulus-response relationships of multistate neural systems

Given recent experimental results suggesting that neural circuits may evolve through multiple firing states, we develop a framework for estimating state-dependent neural response properties from spike train data. We modify the traditional hidden Markov model (HMM) framework to incorporate stimulus-driven, non-Poisson point-process observations. For maximal flexibility, we allow external, time-varying stimuli and the neurons' own spike histories to drive both the spiking behavior in each state and the transitioning behavior between states. We employ an appropriately modified expectation-maximization algorithm to estimate the model parameters. The expectation step is solved by the standard forward-backward algorithm for HMMs. The maximization step reduces to a set of separable concave optimization problems if the model is restricted slightly. We first test our algorithm on simulated data and are able to fully recover the parameters used to generate the data and accurately recapitulate the sequence of hidden states. We then apply our algorithm to a recently published data set in which the observed neuronal ensembles displayed multistate behavior and show that inclusion of spike history information significantly improves the fit of the model. Additionally, we show that a simple reformulation of the state space of the underlying Markov chain allows us to implement a hybrid half-multistate, half-histogram model that may be more appropriate for capturing the complexity of certain data sets than either a simple HMM or a simple peristimulus time histogram model alone.     

Accelerating scientific computations with mixed precision algorithms

On modern architectures, the performance of 32-bit operations is often at least twice as fast as the performance of 64-bit operations. By using a combination of 32-bit and 64-bit floating point arithmetic, the performance of many dense and sparse linear algebra algorithms can be significantly enhanced while maintaining the 64-bit accuracy of the resulting solution. The approach presented here can apply not only to conventional processors but also to other technologies such as Field Programmable Gate Arrays (FPGA), Graphical Processing Units (GPU), and the STI Cell BE processor. Results on modern processor architectures and the STI Cell BE are presented.

Program summary

Program title: ITER-REFCatalogue identifier: AECO_v1_0Program summary URL:http://cpc.cs.qub.ac.uk/summaries/AECO_v1_0.htmlProgram obtainable from: CPC Program Library, Queen's University, Belfast, N. IrelandLicensing provisions: Standard CPC licence, http://cpc.cs.qub.ac.uk/licence/licence.htmlNo. of lines in distributed program, including test data, etc.: 7211No. of bytes in distributed program, including test data, etc.: 41 862Distribution format: tar.gzProgramming language: FORTRAN 77Computer: desktop, serverOperating system: Unix/LinuxRAM: 512 MbytesClassification: 4.8External routines: BLAS (optional)Nature of problem: On modern architectures, the performance of 32-bit operations is often at least twice as fast as the performance of 64-bit operations. By using a combination of 32-bit and 64-bit floating point arithmetic, the performance of many dense and sparse linear algebra algorithms can be significantly enhanced while maintaining the 64-bit accuracy of the resulting solution.Solution method: Mixed precision algorithms stem from the observation that, in many cases, a single precision solution of a problem can be refined to the point where double precision accuracy is achieved. A common approach to the solution of linear systems, either dense or sparse, is to perform the LU factorization of the coefficient matrix using Gaussian elimination. First, the coefficient matrix A is factored into the product of a lower triangular matrix L and an upper triangular matrix U. Partial row pivoting is in general used to improve numerical stability resulting in a factorization PA=LU, where P is a permutation matrix. The solution for the system is achieved by first solving Ly=Pb (forward substitution) and then solving Ux=y (backward substitution). Due to round-off errors, the computed solution, x, carries a numerical error magnified by the condition number of the coefficient matrix A. In order to improve the computed solution, an iterative process can be applied, which produces a correction to the computed solution at each iteration, which then yields the method that is commonly known as the iterative refinement algorithm. Provided that the system is not too ill-conditioned, the algorithm produces a solution correct to the working precision.Running time: seconds/minutes     

Minimize the Maximum Duty in Multi-interface Networks

We consider devices equipped with multiple wired or wireless interfaces. By switching of various interfaces, each device might establish several connections. A connection is established when the devices at its endpoints share at least one active interface. Each interface is assumed to require an activation cost. In this paper, we consider two basic networking problems in the field of multi-interface networks. The first one, known as the Coverage problem, requires to establish the connections defined by a network. The second one, known as Connectivity problem, requires to guarantee a connecting path between any pair of nodes of a network. Both are subject to the constraint of keeping as low as possible the maximum cost set of active interfaces at each single node. We study the problems of minimizing the maximum cost set of active interfaces among the nodes of the network in order to cover all the edges in the first case, or to ensure connectivity in the second case. We prove that the Coverage problem is NP-hard for any fixed Δ≥5 and k≥16, with Δ being the maximum degree, and k being the number of different interfaces among the network. We also show that, unless P=NP, the problem cannot be approximated within a factor of ηln?Δ, for a certain constant η. We then provide a general approximation algorithm which guarantees a factor of O((1+b)ln?Δ), with b being a parameter depending on the topology of the input graph. Interestingly, b can be bounded by a constant for many graph classes. Other approximation and exact algorithms for special cases are presented. Concerning the Connectivity problem, we prove that it is NP-hard for any fixed Δ≥3 and k≥10. Also for this problem, the inapproximability result holds, that is, unless P=NP, the problem cannot be approximated within a factor of ηln?Δ, for a certain constant η. We then provide approximation and exact algorithms for the general problem and for special cases, respectively.     

Classification of human brain tumours from MRS data using Discrete Wavelet Transform and Bayesian Neural Networks

The diagnosis of brain tumours is an extremely sensitive and complex clinical task that must rely upon information gathered through non-invasive techniques. One such technique is Magnetic Resonance Spectroscopy. In this task, radiology experts are likely to benefit from the support of computer-based systems built around robust classification processes. In this paper, a Discrete Wavelet Transform procedure was applied to the pre-processing of spectra corresponding to several brain tumour pathologies. This procedure does not alleviate the high dimensionality of the data by itself. For this reason, dimensionality reduction was subsequently implemented using Moving Window with Variance Analysis for feature selection or Principal Component Analysis for feature extraction. The combined method yielded very encouraging results in terms of diagnostic discriminatory binary classification using Bayesian Neural Networks. In most cases, the classification accuracy improved on previously reported results.     

Adaptive local learning techniques for multiple-step-ahead wind speed forecasting

A massive deployment of wind energy in power systems is expected in the near future. However, a still open issue is how to integrate wind generators into existing electrical grids by limiting their side effects on network operations and control. In order to attain this objective, accurate short and medium-term wind speed forecasting is required.This paper discusses and compares a physical (white-box) model (namely a limited-area non hydrostatic model developed by the European consortium for small-scale modeling) with a family of local learning techniques (black-box) for short and medium term forecasting. Also, an original model integrating machine learning techniques with physical knowledge modeling (grey-box) is proposed.A set of experiments on real data collected from a set of meteorological sensors located in the south of Italy supports the methodological analysis and assesses the potential of the different forecasting approaches.     

On the detection of card-sharing traffic through wavelet analysis and Support Vector Machines

In the last years, the interest in methods and techniques for circumventing the security of the available digital video broadcasting systems is continuously increasing. Digital TV providers are struggling to restrict access to their video contents only to authorized users, by deploying more and more sophisticated conditional access systems. At the state-of-the-art, the most significant menace is the card-sharing activity which exploits a known weakness allowing an authorized subscriber to provide access to digital contents to a potentially large group of unauthorized ones connected over a communication network. This is usually realized by using ad hoc customized devices. Detecting the presence of these illegal systems on a network, by recognizing their related traffic is an issue of primary importance. Unfortunately, to avoid the identification of such traffic, payload obfuscation strategies based on encryption are often used, hindering packet inspection techniques.This paper presents a strategy for the detection of card-sharing traffic, empowered by machine-learning-driven traffic classification techniques and based on the natural capability of wavelet analysis to decompose a traffic time series into several component series associated with particular time and frequency scales and hence allowing its observation at different frequency component levels and with different resolutions. These ideas have been used for the proof-of-concept implementation of an SVM-based binary classification scheme that relies only on time regularities of the traffic and not on the packet contents and hence is immune to payload obfuscation techniques.