Rhythm Patterns Interaction - Synchronization Behavior for Human-Robot Joint Action

Interactive behavior among humans is governed by the dynamics of movement synchronization in a variety of repetitive tasks. This requires the interaction partners to perform for example rhythmic limb swinging or even goal-directed arm movements. Inspired by that essential feature of human interaction, we present a novel concept and design methodology to synthesize goal-directed synchronization behavior for robotic agents in repetitive joint action tasks. The agents’ tasks are described by closed movement trajectories and interpreted as limit cycles, for which instantaneous phase variables are derived based on oscillator theory. Events segmenting the trajectories into multiple primitives are introduced as anchoring points for enhanced synchronization modes. Utilizing both continuous phases and discrete events in a unifying view, we design a continuous dynamical process synchronizing the derived modes. Inverse to the derivation of phases, we also address the generation of goal-directed movements from the behavioral dynamics. The developed concept is implemented to an anthropomorphic robot. For evaluation of the concept an experiment is designed and conducted in which the robot performs a prototypical pick-and-place task jointly with human partners. The effectiveness of the designed behavior is successfully evidenced by objective measures of phase and event synchronization. Feedback gathered from the participants of our exploratory study suggests a subjectively pleasant sense of interaction created by the interactive behavior. The results highlight potential applications of the synchronization concept both in motor coordination among robotic agents and in enhanced social interaction between humanoid agents and humans.     

Case study on a self-organizing spiking neural network for robot navigation

This paper presents a Spiking Neural Network (SNN) architecture for mobile robot navigation. The SNN contains 4 layers where dynamic synapses route information to the appropriate neurons in each layer and the neurons are modeled using the Leaky Integrate and Fire (LIF) model. The SNN learns by self-organizing its connectivity as new environmental conditions are experienced and consequently knowledge about its environment is stored in the connectivity. Also a novel feature of the proposed SNN architecture is that it uses working memory, where present and previous sensor states are stored. Results are presented for a wall following application.     

Parasitic Robot System for Waypoint Navigation of Turtle

In research on small mobile robots and biomimetic robots,locomotion ability remains a major issue despite many advances in technology.However,evolution has led to there being many real animals capable of excellent locomotion.This paper presents a "parasitic robot system" whereby locomotion abilities of an animal are applied to a robot task.We chose a turtle as our first host animal and designed a parasitic robot that can perform "operant conditioning".The parasitic robot,which is attached to the turtle,can induce object-tracking behavior of the turtle toward a Light Emitting Diode (LED) and positively reinforce the behavior through repeated stimulus-response interaction.After training sessions over five weeks,the robot could successfully control the direction of movement of the trained turtles in the waypoint navigation task.This hybrid animal-robot interaction system could provide an alternative solution to some of the limitations of conventional mobile robot systems in various fields,and could also act as a useful interaction system for the behavioral sciences.     

Adaptive Neural Network Structure Optimization Algorithm Based on Dynamic Nodes

Large-scale artificial neural networks have many redundant structures, making the network fall into the issue of local optimization and extended training time. Moreover, existing neural network topology optimization algorithms have the disadvantage of many calculations and complex network structure modeling. We propose a Dynamic Node-based neural network Structure optimization algorithm (DNS) to handle these issues. DNS consists of two steps: the generation step and the pruning step. In the generation step, the network generates hidden layers layer by layer until accuracy reaches the threshold. Then, the network uses a pruning algorithm based on Hebb’s rule or Pearson’s correlation for adaptation in the pruning step. In addition, we combine genetic algorithm to optimize DNS (GA-DNS). Experimental results show that compared with traditional neural network topology optimization algorithms, GA-DNS can generate neural networks with higher construction efficiency, lower structure complexity, and higher classification accuracy.     

Artificial neural networks for blur identification and restoration of nonlinearly degraded images.

In this paper, an image restoration algorithm is proposed to identify noncausal blur function. Image degradation processes include both linear and nonlinear phenomena. A neural network model combining an adaptive auto-associative network with a random Gaussian process is proposed to restore the blurred image and blur function simultaneously. The noisy and blurred images are modeled as continuous associative networks, whereas auto-associative part determines the image model coefficients and the hetero-associative part determines the blur function of the system. The self-organization like structure provides the potential solution of the blind image restoration problem. The estimation and restoration are implemented by using an iterative gradient based algorithm to minimize the error function.     

Adaptive Robotic Control Driven by a Versatile Spiking Cerebellar Network

The cerebellum is involved in a large number of different neural processes, especially in associative learning and in fine motor control. To develop a comprehensive theory of sensorimotor learning and control, it is crucial to determine the neural basis of coding and plasticity embedded into the cerebellar neural circuit and how they are translated into behavioral outcomes in learning paradigms. Learning has to be inferred from the interaction of an embodied system with its real environment, and the same cerebellar principles derived from cell physiology have to be able to drive a variety of tasks of different nature, calling for complex timing and movement patterns. We have coupled a realistic cerebellar spiking neural network (SNN) with a real robot and challenged it in multiple diverse sensorimotor tasks. Encoding and decoding strategies based on neuronal firing rates were applied. Adaptive motor control protocols with acquisition and extinction phases have been designed and tested, including an associative Pavlovian task (Eye blinking classical conditioning), a vestibulo-ocular task and a perturbed arm reaching task operating in closed-loop. The SNN processed in real-time mossy fiber inputs as arbitrary contextual signals, irrespective of whether they conveyed a tone, a vestibular stimulus or the position of a limb. A bidirectional long-term plasticity rule implemented at parallel fibers-Purkinje cell synapses modulated the output activity in the deep cerebellar nuclei. In all tasks, the neurorobot learned to adjust timing and gain of the motor responses by tuning its output discharge. It succeeded in reproducing how human biological systems acquire, extinguish and express knowledge of a noisy and changing world. By varying stimuli and perturbations patterns, real-time control robustness and generalizability were validated. The implicit spiking dynamics of the cerebellar model fulfill timing, prediction and learning functions.     

Application of chaotic dynamics in a recurrent neural network to control: hardware implementation into a novel autonomous roving robot

Originating from a viewpoint that complex/chaotic dynamics would play an important role in biological system including brains, chaotic dynamics introduced in a recurrent neural network was applied to control. The results of computer experiment was successfully implemented into a novel autonomous roving robot, which can only catch rough target information with uncertainty by a few sensors. It was employed to solve practical two-dimensional mazes using adaptive neural dynamics generated by the recurrent neural network in which four prototype simple motions are embedded. Adaptive switching of a system parameter in the neural network results in stationary motion or chaotic motion depending on dynamical situations. The results of hardware implementation and practical experiment using it show that, in given two-dimensional mazes, the robot can successfully avoid obstacles and reach the target. Therefore, we believe that chaotic dynamics has novel potential capability in controlling, and could be utilized to practical engineering application.     

Formal Models of the Network Co-occurrence Underlying Mental Operations

Systems neuroscience has identified a set of canonical large-scale networks in humans. These have predominantly been characterized by resting-state analyses of the task-unconstrained, mind-wandering brain. Their explicit relationship to defined task performance is largely unknown and remains challenging. The present work contributes a multivariate statistical learning approach that can extract the major brain networks and quantify their configuration during various psychological tasks. The method is validated in two extensive datasets (n = 500 and n = 81) by model-based generation of synthetic activity maps from recombination of shared network topographies. To study a use case, we formally revisited the poorly understood difference between neural activity underlying idling versus goal-directed behavior. We demonstrate that task-specific neural activity patterns can be explained by plausible combinations of resting-state networks. The possibility of decomposing a mental task into the relative contributions of major brain networks, the "network co-occurrence architecture" of a given task, opens an alternative access to the neural substrates of human cognition.     

Motion planning with complete knowledge using a colored SOM

The motion planning problem requires that a collision-free path be determined for a robot moving amidst a fixed set of obstacles. Most neural network approaches to this problem are for the situation in which only local knowledge about the configuration space is available. The main goal of the paper is to show that neural networks are also suitable tools in situations with complete knowledge of the configuration space. In this paper we present an approach that combines a neural network and deterministic techniques. We define a colored version of Kohonen's self-organizing map that consists of two different classes of nodes. The network is presented with random configurations of the robot and, from this information, it constructs a road map of possible motions in the work space. The map is a growing network, and different nodes are used to approximate boundaries of obstacles and the Voronoi diagram of the obstacles, respectively. In a second phase, the positions of the two kinds of nodes are combined to obtain the road map. In this way a number of typical problems with small obstacles and passages are avoided, and the required number of nodes for a given accuracy is within reasonable limits. This road map is searched to find a motion connecting the given source and goal configurations of the robot. The algorithm is simple and general; the only specific computation that is required is a check for intersection of two polygons. We implemented the algorithm for planar robots allowing both translation and rotation and experiments show that compared to conventional techniques it performs well, even for difficult motion planning scenes.     

A hierarchical method for finding optimal architecture and weights using evolutionary least square based learning

In this paper, we present a novel approach of implementing a combination methodology to find appropriate neural network architecture and weights using an evolutionary least square based algorithm (GALS).1 This paper focuses on aspects such as the heuristics of updating weights using an evolutionary least square based algorithm, finding the number of hidden neurons for a two layer feed forward neural network, the stopping criterion for the algorithm and finally some comparisons of the results with other existing methods for searching optimal or near optimal solution in the multidimensional complex search space comprising the architecture and the weight variables. We explain how the weight updating algorithm using evolutionary least square based approach can be combined with the growing architecture model to find the optimum number of hidden neurons. We also discuss the issues of finding a probabilistic solution space as a starting point for the least square method and address the problems involving fitness breaking. We apply the proposed approach to XOR problem, 10 bit odd parity problem and many real-world benchmark data sets such as handwriting data set from CEDAR, breast cancer and heart disease data sets from UCI ML repository. The comparative results based on classification accuracy and the time complexity are discussed.     

Hierarchical genetic algorithm for near optimal feedforward neural network design

In this paper, we propose a genetic algorithm based design procedure for a multi layer feed forward neural network. A hierarchical genetic algorithm is used to evolve both the neural networks topology and weighting parameters. Compared with traditional genetic algorithm based designs for neural networks, the hierarchical approach addresses several deficiencies, including a feasibility check highlighted in literature. A multi objective cost function is used herein to optimize the performance and topology of the evolved neural network simultaneously. In the prediction of Mackey Glass chaotic time series, the networks designed by the proposed approach prove to be competitive, or even superior, to traditional learning algorithms for the multi layer Perceptron networks and radial basis function networks. Based upon the chosen cost function, a linear weight combination decision making approach has been applied to derive an approximated Pareto optimal solution set. Therefore, designing a set of neural networks can be considered as solving a two objective optimization problem.     

Evolvable neuronal paths: a novel basis for information and search in the brain

We propose a previously unrecognized kind of informational entity in the brain that is capable of acting as the basis for unlimited hereditary variation in neuronal networks. This unit is a path of activity through a network of neurons, analogous to a path taken through a hidden Markov model. To prove in principle the capabilities of this new kind of informational substrate, we show how a population of paths can be used as the hereditary material for a neuronally implemented genetic algorithm, (the swiss-army knife of black-box optimization techniques) which we have proposed elsewhere could operate at somatic timescales in the brain. We compare this to the same genetic algorithm that uses a standard 'genetic' informational substrate, i.e. non-overlapping discrete genotypes, on a range of optimization problems. A path evolution algorithm (PEA) is defined as any algorithm that implements natural selection of paths in a network substrate. A PEA is a previously unrecognized type of natural selection that is well suited for implementation by biological neuronal networks with structural plasticity. The important similarities and differences between a standard genetic algorithm and a PEA are considered. Whilst most experiments are conducted on an abstract network model, at the conclusion of the paper a slightly more realistic neuronal implementation of a PEA is outlined based on Izhikevich spiking neurons. Finally, experimental predictions are made for the identification of such informational paths in the brain.     

A fuzzy-based reliability system for knowledge sharing between robots in P2P JXTA-overlay platform

The design of an efficient collaborative multi-robot framework that ensures the autonomy and the individual requirements of the involved robots is a very challenging task. This requires designing an efficient platform for inter-robot communication. P2P is a good approach to achieve this goal. P2P aims at making the communication ubiquitous thereby crossing the communication boundary and has many attractive features to use it as a platform for collaborative multi-robot environments. In this paper, we present our implemented P2P system based on JXTA Overlay. We use JXTA Overlay as a platform for robot collaboration and knowledge sharing. We also propose a fuzzy-based peer reliability system for JXTA-Overlay platform considering three parameters: Actual Behavior Criterion (ABC), Mutually Agreed Behavior (MAB) and Reputation (R). We evaluated the knowledge sharing system by many experiments and show that this system has a good performance and can be used successfully for knowledge sharing between robots. Also, we present some simulation results, which show the fuzzy-based peer reliability system has a good behavior and can successfully select the best peer candidate.     

A hidden Markov model for continuous longitudinal data with missing responses and dropout

We propose a hidden Markov model for multivariate continuous longitudinal responses with covariates that accounts for three different types of missing pattern: (I) partially missing outcomes at a given time occasion, (II) completely missing outcomes at a given time occasion (intermittent pattern), and (III) dropout before the end of the period of observation (monotone pattern). The missing-at-random (MAR) assumption is formulated to deal with the first two types of missingness, while to account for the informative dropout, we rely on an extra absorbing state. Estimation of the model parameters is based on the maximum likelihood method that is implemented by an expectation-maximization (EM) algorithm relying on suitable recursions. The proposal is illustrated by a Monte Carlo simulation study and an application based on historical data on primary biliary cholangitis.     

An approximation algorithm for haplotype inference by maximum parsimony.

This paper studies haplotype inference by maximum parsimony using population data. We define the optimal haplotype inference (OHI) problem as given a set of genotypes and a set of related haplotypes, find a minimum subset of haplotypes that can resolve all the genotypes. We prove that OHI is NP-hard and can be formulated as an integer quadratic programming (IQP) problem. To solve the IQP problem, we propose an iterative semidefinite programming-based approximation algorithm, (called SDPHapInfer). We show that this algorithm finds a solution within a factor of O(log n) of the optimal solution, where n is the number of genotypes. This algorithm has been implemented and tested on a variety of simulated and biological data. In comparison with three other methods, (1) HAPAR, which was implemented based on the branching and bound algorithm, (2) HAPLOTYPER, which was implemented based on the expectation-maximization algorithm, and (3) PHASE, which combined the Gibbs sampling algorithm with an approximate coalescent prior, the experimental results indicate that SDPHapInfer and HAPLOTYPER have similar error rates. In addition, the results generated by PHASE have lower error rates on some data but higher error rates on others. The error rates of HAPAR are higher than the others on biological data. In terms of efficiency, SDPHapInfer, HAPLOTYPER, and PHASE output a solution in a stable and consistent way, and they run much faster than HAPAR when the number of genotypes becomes large.     

Cooperative coevolution of neural representations

A genetic algorithm (GA) is used to search for a set of local feature detectors or hidden units. These are in turn employed as a representation of the input data for neural learning in the upper layer of a multilayer perceptron (MLP) which performs an image classification task. Three different methods of encoding hidden unit weights in the chromosome of the GA are presented, including one which coevolves all the feature detectors in a single chromosome, and two which promote the cooperation of feature detectors by encoding them in their own individual chromosomes. The fitness function measures the MLP classification accuracy together with the confidence of the networks.     

Emergence of functional hierarchy in a multiple timescale neural network model: a humanoid robot experiment

It is generally thought that skilled behavior in human beings results from a functional hierarchy of the motor control system, within which reusable motor primitives are flexibly integrated into various sensori-motor sequence patterns. The underlying neural mechanisms governing the way in which continuous sensori-motor flows are segmented into primitives and the way in which series of primitives are integrated into various behavior sequences have, however, not yet been clarified. In earlier studies, this functional hierarchy has been realized through the use of explicit hierarchical structure, with local modules representing motor primitives in the lower level and a higher module representing sequences of primitives switched via additional mechanisms such as gate-selecting. When sequences contain similarities and overlap, however, a conflict arises in such earlier models between generalization and segmentation, induced by this separated modular structure. To address this issue, we propose a different type of neural network model. The current model neither makes use of separate local modules to represent primitives nor introduces explicit hierarchical structure. Rather than forcing architectural hierarchy onto the system, functional hierarchy emerges through a form of self-organization that is based on two distinct types of neurons, each with different time properties ("multiple timescales"). Through the introduction of multiple timescales, continuous sequences of behavior are segmented into reusable primitives, and the primitives, in turn, are flexibly integrated into novel sequences. In experiments, the proposed network model, coordinating the physical body of a humanoid robot through high-dimensional sensori-motor control, also successfully situated itself within a physical environment. Our results suggest that it is not only the spatial connections between neurons but also the timescales of neural activity that act as important mechanisms leading to functional hierarchy in neural systems.     

How to reconstruct a large genetic network from n gene perturbations in fewer than n(2) easy steps.

MOTIVATION: The reconstruction of genetic networks is the holy grail of functional genomics. Its core task is to identify the causal structure of a gene network, that is, to distinguish direct from indirect regulatory interactions among gene products. In other words, to reconstruct a genetic network is to identify, for each network gene, which other genes and their activity the gene influences directly. Crucial to this task are perturbations of gene activity. Genomic technology permits large-scale experiments perturbing the activity of many genes and assessing the effect of each perturbation on all other genes in a genome. However, such experiments cannot distinguish between direct and indirect effects of a genetic perturbation. RESULTS: I present an algorithm to reconstruct direct regulatory interactions in gene networks from the results of gene perturbation experiments. The algorithm is based on a graph representation of genetic networks and applies to networks of arbitrary size and complexity. Algorithmic complexity in both storage and time is low, less than O(n(2)). In practice, the algorithm can reconstruct networks of several thousand genes in mere CPU seconds on a desktop workstation. AVAILABILITY: A perl implementation of the algorithm is given in the Appendix. CONTACT: wagnera@unm.edu     

Complementary Roles of the Hippocampus and the Dorsomedial Striatum during Spatial and Sequence-Based Navigation Behavior

We investigated the neural bases of navigation based on spatial or sequential egocentric representation during the completion of the starmaze, a complex goal-directed navigation task. In this maze, mice had to swim along a path composed of three choice points to find a hidden platform. As reported previously, this task can be solved by using two hippocampal-dependent strategies encoded in parallel i) the allocentric strategy requiring encoding of the contextual information, and ii) the sequential egocentric strategy requiring temporal encoding of a sequence of successive body movements associated to specific choice points. Mice were trained during one day and tested the following day in a single probe trial to reveal which of the two strategies was spontaneously preferred by each animal. Imaging of the activity-dependent gene c-fos revealed that both strategies are supported by an overlapping network involving the dorsal hippocampus, the dorsomedial striatum (DMS) and the medial prefrontal cortex. A significant higher activation of the ventral CA1 subregion was observed when mice used the sequential egocentric strategy. To investigate the potential different roles of the dorsal hippocampus and the DMS in both types of navigation, we performed region-specific excitotoxic lesions of each of these two structures. Dorsal hippocampus lesioned mice were unable to optimally learn the sequence but improved their performances by developing a serial strategy instead. DMS lesioned mice were severely impaired, failing to learn the task. Our data support the view that the hippocampus organizes information into a spatio-temporal representation, which can then be used by the DMS to perform goal-directed navigation.