Ecological Neural Networks for Object Recognition and Generalization

Generalization is a critical capacity for organisms. Modeling the behavior of organisms with neural networks, some type of generalizations appear to be accessible to neural networks but other types do not. In this paper we present two simulations. In the first simulation we show that while neural networks can recognize where an object is located in the retina even if they have never experienced that object in that position ('where' generalization subtask), they have difficulty in recognizing the identity of a familiar object in a new position ('what' generalization subtask). In the second simulation we explore the hypothesis that organisms find another solution to the problem of recognizing objects in different positions on their retina: they move their eyes so that objects are always seen in the same position in the retina. This strategy emerges spontaneously in ecological neural networks that are allowed to move their 'eye' in order to bring different portions of the visible world in the central portion of their retina.     

The Tenet experience and the design of protocols for integrated-services internetworks

The research done by the Tenet Group in multimedia networking has reached a point where it may be useful to reflect on the significance of its results for the current debate on how integrated-services internetworks should be designed. Such reflections constitute the main subject of this paper. The principles of the work and the conclusions reached so far by the Tenet researchers are discussed in the light of the conflict between the two major technologies being proposed to build future information infrastructures: namely, the Internet and the ATM technologies. The Tenet approach suggests one feasible way for resolving the conflict to the advantage of all the users of those infrastructures. This paper discusses various fundamental aspects of integrated-services network design: the choice of the service model, the type of charging policy to be adopted, and the selection of a suitable architecture.     

Good teaching inputs do not correspond to desired responses in ecological neural networks

In this paper we analyze how supervised learning occurs in ecological neural networks, i.e. networks that interact with an autonomous external environment and, therefore, at least partially determine with their behavior their own input. Using an evolutionary method for selecting good teaching inputs we surprisingly find that to obtain a desired outputX it is better to use a teaching input different fromX. To explain this fact we claim that teaching inputs in ecological networks have two different effects: (a) to reduce the discrepancy between the actual output of the network and the teaching input, (b) to modify the network's behavior and, as a consequence, the network's learning experiences. Evolved teaching inputs appear to represent a compromise between these two needs. We finally show that evolved teaching inputs that are allowed to change during the learning process function differently at different stages of learning, first giving more weight to (b) and, later on, to (a).     

Convolution profiles for right inversion of multivariable non-minimum phase discrete-time systems

The problem of the non-causal inversion of linear multivariable discrete-time systems is analyzed in the geometric approach framework and is solved through the computation of convolution profiles which guarantee perfect tracking under the assumption of infinite-length preaction and postaction time intervals. It is shown how the shape of the convolution profiles is related to both the relative degree and the invariant zeros of the plant. A computational setting for the convolution profiles is derived by means of the standard geometric approach tools. Feasibility constraints are also taken into account. A possible implementation scheme, based on a finite impulse response system acting on a stabilized control loop, is provided.     

Compared Analysis of Metro Networks Supported by Graph Theory

This paper brings into focus the topological and geographical evaluation of metro networks through the definition of a methodological approach based on a set of indicators, a lot of which are defined in the sector literature. Once the methodology was stated, the results of an application on the metro networks of 13 big metropolitan areas were illustrated. Statistical comparative analyses are proposed to classify networks.     

Input-Output Equivalence of Spin Networks under Multiple Measurements

Two quantum control systems that are driven by an external field are said to be input–output equivalent if, for any control field, the measured value of a given observable is the same. Equivalent models cannot be distinguished by experiments involving state evolutions and measurements. In this paper, we characterize the equivalent models of networks of spin driven by electro-magnetic fields for which the expectation value of the total magnetization is measured. Extending previous results and definitions that only dealt with the case of a single measurement, we describe the class of equivalent models under a sequence of Von Neumann measurements. The results are motivated by the problem of parameter identification for Heisenberg spin systems modeling molecular magnets.Mathematics Subject Classifications (2000): 93B30, 17B45, 17B81Partially supported by M.I.U.R., project Viscosity, metric, and control theoretic methods for nonlinear partial differential equations.Supported by NSF Career Grant, ECS0237925.     

Two is better than one: A diploid genotype for neural networks

In nature the genotype of many organisms exhibits diploidy, i.e., it includes two copies of every gene. In this paper we describe the results of simulations comparing the behavior of haploid and diploid populations of ecological neural networks living in both fixed and changing environments. We show that diploid genotypes create more variability in fitness in the population than haploid genotypes and buffer better environmental change; as a consequence, if one wants to obtain good results for both average and peak fitness in a single population one should choose a diploid population with an appropriate mutation rate. Some results of our simulations parallel biological findings.     

Optimal evaluation of generalized Euler angles with applications to control

Given two linearly independent matrices in so(3), Z₁ and Z₂, every rotation matrix, X_f∈SO(3), can be written as the product of alternate elements from the one-dimensional subgroups corresponding to Z₁ and Z₂, namely X_f=e^Z₁t₁e^Z₂t₂e^Z₁t₃?e^Z₁t_s. The parameters t_i, i=1,…,s are called Generalized Euler Angles. In this paper, the minimum number of factors required for the factorization of X_f∈SO(3), as a function of X_f, is evaluated. An algorithm is given to determine the generalized Euler angles, in the optimal factorization. The results can be applied to the bang-bang control, with minimum number of switches, of some classical and quantum systems.