T. Martin McGinnity

Advances in Design and Application of Spiking Neural Networks

This paper presents new findings in the design and application of biologically plausible neural networks based on spiking neuron models, which represent a more plausible model of real biological neurons where time is considered as an important feature for information encoding and processing in the brain. The design approach consists of an evolutionary strategy based supervised training algorithm, newly developed by the authors, and the use of different biologically plausible neuronal models. A dynamic synapse (DS) based neuron model, a biologically more detailed model, and the spike response model (SRM) are investigated in order to demonstrate the efficacy of the proposed approach and to further our understanding of the computing capabilities of the nervous system. Unlike the conventional synapse, represented as a static entity with a fixed weight, employed in conventional and SRM-based neural networks, a DS is weightless and its strength changes upon the arrival of incoming input spikes. Therefore its efficacy depends on the temporal structure of the impinging spike trains. In the proposed approach, the training of the network free parameters is achieved using an evolutionary strategy where, instead of binary encoding, real values are used to encode the static and DS parameters which underlie the learning process. The results show that spiking neural networks based on both types of synapse are capable of learning non-linearly separable data by means of spatio-temporal encoding. Furthermore, a comparison of the obtained performance with classical neural networks (multi-layer perceptrons) is presented.

Edge Detection Based on Spiking Neural Network Model

Inspired by the behaviour of biological receptive fields and the human visual system, a network model based on spiking neurons is proposed to detect edges in a visual image. The structure and the properties of the network are detailed in this paper. Simulation results show that the network based on spiking neurons is able to perform edge detection within a time interval of 100 ms. This processing time is consistent with the human visual system. A firing rate map recorded in the simulation is comparable to Sobel and Canny edge graphics. In addition, the network can separate different edges using synapse plasticity, and the network provides an attention mechanism in which edges in an attention area can be enhanced.

Image-Based Visual Servoing of a 7-DOF Robot Manipulator Using an Adaptive Distributed Fuzzy PD Controller

This paper is concerned with the design and implementation of a distributed proportional-derivative (PD) controller of a 7-degrees of freedom (DOF) robot manipulator using the Takagi-Sugeno (T-S) fuzzy framework. Existing machine learning approaches to visual servoing involve system identification of image and kinematic Jacobians. In contrast, the proposed approach actuates a control signal primarily as a function of the error and derivative of the error in the desired visual feature space. This approach leads to a significant reduction in the computational burden as compared to model-based approaches, as well as existing learning approaches to model inverse kinematics. The simplicity of the controller structure will make it attractive in industrial implementations where PD/PID type schemes are in common use. While the initial values of PD gain are learned with the help of model-based controller, an online adaptation scheme has been proposed that is capable of compensating for local uncertainties associated with the system and its environment. Rigorous experiments have been performed to show that visual servoing tasks such as reaching a static target and tracking of a moving target can be achieved using the proposed distributed PD controller. It is shown that the proposed adaptive scheme can dynamically tune the controller parameters during visual servoing, so as to improve its initial performance based on parameters obtained while mimicking the model-based controller. The proposed control scheme is applied and assessed in real-time experiments using an uncalibrated eye-in-hand robotic system with a 7-DOF PowerCube robot manipulator.

Learning under weight constraints in networks of temporal encoding spiking neurons

Limits on synaptic efficiency are characteristic of biological neural networks. In this paper, weight limitation constraints are applied to the spike time error-backpropagation (SpikeProp) algorithm for temporally encoded networks of spiking neurons. A novel solution to the problem raised by non-firing neurons is presented which makes the learning algorithm converge reliably and efficiently. In addition a square cosine encoder is applied to the input neurons to reduce the number of input neurons required. The approach is demonstrated by application to the classical XOR-problem analysis, a function approximation experiment and benchmark data sets. Using input delay neurons and relative timing, the algorithm is also applied to solve a time series prediction problem. The experimental results show that the new approach produces comparable accuracy in classification with the original approach while utilising a smaller spiking neural network.

A voxel based morphometry study investigating brain structural changes in first episode psychosis.

Schizophrenia (SCZ) and bipolar disorder (BP) are associated with neuropathological brain changes, which are believed to disrupt connectivity between brain processes and may have common properties. Patients at first psychotic episode are unique, as one can assess brain alterations at illness inception, when many confounders are reduced or absent. SCZ (N=25) and BP (N=24) patients were recruited in a regional first episode psychosis MRI study. VBM methods were used to study gray matter (GM) and white matter (WM) differences between patient groups and case by case matched controls. For both groups, deficits identified are more discrete than those typically reported in later stages of illness. SCZ patients showed some evidence of GM loss in cortical areas but most notable were in limbic structures such as hippocampus, thalamus and striatum and cerebellum. Consistent with disturbed neural connectivity WM alterations were also observed in limbic structures, the corpus callosum and many subgyral and sublobar regions in the parietal, temporal and frontal lobes. BP patients displayed less evidence of volume changes overall, compared to normal healthy participants, but those changes observed were primarily in WM areas which overlapped with regions identified in SCZ, including thalamus and cerebellum and subgyral and sublobar sites. At first episode of psychosis there is evidence of a neuroanatomical overlap between SCZ and BP with respect to brain structural changes, consistent with disturbed neural connectivity. There are also important differences however in that SCZ displays more extensive structural alteration.

EEG-based continuous control of a game using a 3 channel motor imagery BCI: BCI game

This paper presents an overview of a multistage signal processing framework to tackle the main challenges in continuous control protocols for motor imagery based synchronous and self-paced BCIs. The BCI can be setup rapidly and automatically even when conducting an extensive search for subject-specific parameters. A new BCI-based game training paradigm which enables assessment of continuous control performance is also introduced. A range of offline results and online analysis of the new game illustrate the potential for the proposed BCI and the advantages of using the game as a BCI training paradigm.

Multiknowledge for decision making

The representation of knowledge has an important effect on automated decision-making. In this paper, vector spaces are used to describe a condition space and a decision space, and knowledge is represented by a mapping from the condition space to the decision space. Many such mappings can be obtained from a training set. A set of mappings, which are created from multiple reducts in the training set, is defined as multiknowledge. In order to get a good reduct and find multiple reducts, the WADF (worst-attribute-drop-first) algorithm is developed through analysis of the properties of decision systems using rough set theory. An approach that combines multiknowledge and the naïve Bayes classifier is applied to make decisions for unseen instances or for instances with missing attribute values. Benchmark data sets from the UCI Machine Learning Repository are used to test the algorithms. The experimental results are encouraging; the prediction accuracy for unseen instances by using the algorithms is higher than by using other approaches based on a single body of knowledge.

Robotic UBIquitous COgnitive Network

Robotic ecologies are networks of heterogeneous robotic devices pervasively embedded in everyday environments, where they cooperate to perform complex tasks. While their potential makes them increasingly popular, one fundamental problem is how to make them self-adaptive, so as to reduce the amount of preparation, pre-programming and human supervision that they require in real world applications. The EU FP7 project RUBICON develops self-sustaining learning solutions yielding cheaper, adaptive and efficient coordination of robotic ecologies. The approach we pursue builds upon a unique combination of methods from cognitive robotics, agent control systems, wireless sensor networks and machine learning. This paper briefly illustrates how these techniques are being extended, integrated, and applied to AAL applications.

AN STDP TRAINING ALGORITHM FOR A SPIKING NEURAL NETWORK WITH DYNAMIC THRESHOLD NEURONS

This paper proposes a supervised training algorithm for Spiking Neural Networks (SNNs) which modifies the Spike Timing Dependent Plasticity (STDP)learning rule to support both local and network level training with multiple synaptic connections and axonal delays. The training algorithm applies the rule to two and three layer SNNs, and is benchmarked using the Iris and Wisconsin Breast Cancer (WBC) data sets. The effectiveness of hidden layer dynamic threshold neurons is also investigated and results are presented.

The sound-induced flash illusion reveals dissociable age-related effects in multisensory integration

While aging can lead to significant declines in perceptual and cognitive function, the effects of age on multisensory integration, the process in which the brain combines information across the senses, are less clear. Recent reports suggest that older adults are susceptible to the sound-induced flash illusion (Shams et al., 2000) across a much wider range of temporal asynchronies than younger adults (Setti et al., 2011). To assess whether this cost for multisensory integration is a general phenomenon of combining asynchronous audiovisual input, we compared the time courses of two variants of the sound-induced flash illusion in young and older adults: the fission illusion, where one flash accompanied by two beeps appears as two flashes, and the fusion illusion, where two flashes accompanied by one beep appear as one flash. Twenty-five younger (18–30 years) and older (65+ years) adults were required to report whether they perceived one or two flashes, whilst ignoring irrelevant auditory beeps, in bimodal trials where auditory and visual stimuli were separated by one of six stimulus onset asynchronies (SOAs). There was a marked difference in the pattern of results for the two variants of the illusion. In conditions known to produce the fission illusion, older adults were significantly more susceptible to the illusion at longer SOAs compared to younger participants. In contrast, the performance of the younger and older groups was almost identical in conditions known to produce the fusion illusion. This surprising difference between sound-induced fission and fusion in older adults suggests dissociable age-related effects in multisensory integration, consistent with the idea that these illusions are mediated by distinct neural mechanisms.