Fig 21 - uploaded by Alessandro E. P. Villa
Content may be subject to copyright.
Source publication
One of the major obstacles found when trying to construct artefacts derived from principles observed in living beings is the lack of actual dynamic hardware with autonomous capabilities. Even if programmable devices offer the possibility of modifying the functionality implemented in the device, they rely on external hardware and software elements t...
Context in source publication
Similar publications
This paper presents a novel real-time dynamic framework for quantifying time-series structure in spoken words using spikes. Audio signals are converted into multi-channel spike trains using a biologically-inspired leaky integrate-and-fire (LIF) spike generator. These spike trains are mapped into a function space of infinite dimension, i.e., a Repro...
Defects pose a significant challenge to the reliability of electronic devices, particularly when dealing with scaled dimensions in the silicon microelectronic industry. Consequently, extensive efforts have been made to eliminate these defects from devices through the optimization of growth and fabrication processes. However, in the realm of emergin...
Learning and development in real brains typically happens over long timescales, making long-term exploration of these features a significant research challenge. One way to address this problem is to use computational models to explore the brain, with Spiking Neural Networks a popular choice to capture neuron and synapse dynamics. However, researche...
One of the directions of development within the framework of the neuromorphic approach is the development of anatomically similar models of brain networks, taking into account the structurally complex structure of neurons and the adaptation of connections between them, as well as the development of learning algorithms for such models. In this work,...
![Figure 5: The same as in Fig. 4 but with spike amplitudes in [−3/2ϑ, 3/2ϑ].](publication/370776120/figure/fig4/AS:11431281158666557@1684206786089/The-same-as-in-Fig-4-but-with-spike-amplitudes-in-3-2th-3-2th_Q320.jpg)
In spiking neural networks (SNN), at each node, an incoming sequence of weighted Dirac pulses is converted into an output sequence of weighted Dirac pulses by a leaky-integrate-and-fire (LIF) neuron model based on spike aggregation and thresholding. We show that this mapping can be understood as a quantization operator and state a corresponding for...
Citations
... 27 This electronic circuit is a flexible hardware substrate showing the basic features that permit living beings to show evolutionary, developmental or learning capabilities. 28 In future work, these features are intended to be implemented into a novel and even more flexible hardware architecture called ubidule a. The genomic features of these hardware tissues offer the possibility to implement programmed cell death mechanisms in simulations of large spiking neural networks. ...
Two main processes concurrently refine the nervous system over the course of development: cell death and selective synaptic pruning. We simulated large spiking neural networks (100 x 100 neurons "at birth") characterized by an early developmental phase with cell death due to excessive firing rate, followed by the onset of spike timing dependent synaptic plasticity (STDP), driven by spatiotemporal patterns of stimulation. The cell death affected the inhibitory units more than the excitatory units during the early developmental phase. The network activity showed the appearance of recurrent spatiotemporal firing patterns along the STDP phase, thus suggesting the emergence of cell assemblies from the initially randomly connected networks. Some of these patterns were detected throughout the simulation despite the activity-driven network modifications while others disappeared.
... These are called POE systems as they are aimed to show these capabilities in all three aspects of Phylogeny, Ontogeny and, Epigenesis of an organism. A spiking neural network on the POEtic chip [26] is an example of such systems. ...
... Thus, a time-step simulation technique is used here. This model also needs to be relatively fast as running a POE system [26] involves iterative nested cycles of evolution, development and learning. Such a fast parallel spiking neural network on FPGA can also be used for real-time applications. ...
... Compared to existing hardware-based models (e.g. [30], [35], [25], [26]), the new digital neuron model has many advantages: ...
We propose a digital neuron model suitable for evolving and growing heterogeneous spiking neural networks on FPGAs by introducing a novel flexible dendrite architecture and the new PLAQIF (piecewise-linear approximation of quadratic integrate and fire) soma model. A network of 161 neurons and 1610 synapses was simulated, implemented, and verified on a Virtex-5 chip with 4210 times real-time speed with 1 ms resolution. The parametric flexibility of the soma model was shown through a set of experiments.
... Researchers have sought to create POE (Phylogeny, Ontogeny, Epigenesis) neural networks capable of evolving, de- veloping (growth) and learning in situ to adapt themselves to the given problem and environment (e.g. POEtic chip [5]). However, as yet none of these digital neuron models are quite suitable for an online developmental model [1] capable of regeneration and growth (Morphogenesis) on FPGA. ...
... Thus, a time-step simulation technique is used here. This model also needs to be relatively fast as running a POE system [5] involves iter- ative nested cycles of evolution, development and learning. Such a fast parallel spiking neural network on FPGA can also be used for real-time applications. ...
We propose a digital neuron model suitable for evolving and growing heterogeneous spiking neural networks on FPGAs using a piecewise linear approximation of the Quadratic Integrate and Fire (QIF) model. A network of 161 neurons and 1610 synapses with 4210 times real- time neuron simulation speed was simulated and synthesized for a Virtex-5 chip.
Because of the human brain’s parallel computing structure and its characteristics of the localized storage, the human brain has great superiority of high throughput and low power consumption. Based on the bionics of the brain, many researchers try to imitate the behavior of neurons with hardware platform so that we can obtain the same or close computational acceleration performance like the brain. In this paper, we proposed a hardware structure to implement single neuron with Integration-and-Fire(IF) model on Virtex-7 XC7VX485T-ffg1157 FPGA. Through simulation and synthesis, we quantitatively analyzed the device utilization and power consumption of our structure; meanwhile, the function of the proposed hardware implementation is verified with the classic XOR benchmark with a 4-layer SNN and the scalability of our hardware neuron is tested with a handwritten digits recognition mission on MNIST database using a 6-layer SNN. Experimental results show that the neuron hardware implementation proposed in this paper can pass the XOR benchmark test and fulfill the need of handwritten digits recognition mission. The total on-chip power of 4-layer SNN is 0.114 W, which is the lowest among the ANN and firing-rate based SNN at the same scale.
In this paper we shall review the major approaches done in the past concerning the development of immersive artistic installations. This will permit one to identify the basic features to be attained by this kind of installations and based on them a new cooperative and autonomous robotic platform will be proposed. The major novelty of the approach presented in this paper relies on its short- and long-term adaptation capabilities. The technical details of its basic building blocks and the results of the preliminary field tests will be provided, showing that it constitutes a very efficient platform for the construction of actual adaptive and immersive artistic installations.
A scalable hardware platform made of custom reconfigurable devices endowed with bio-inspired ontogenetic and epigenetic features
is configured to run an artificial neural network with developmental and evolvable capabilities. The hardware architecture
allows internetwork communication and this study analyzes the simulated activity of two hierarchically organized spiking neural
networks. The main features were an initial developmental phase characterized by cell death (apoptosis driven by excessive
firing rate), followed by spike timing dependent synaptic plasticity in presence of background noise. The emergence of precise
firing sequences formed by recurrent patterns of spike intervals above chance levels suggested the build-up of a connectivity,
out of initially randomly connected networks, able to sustain temporal information processing. The relative frequency of precise
firing sequences was higher in the downstream network and their dynamics suggested the emergence of an unsupervised hierarchical
activity-driven connectivity.
Functional magnetic resonance imaging (fMRI) triggered by scalp electroencephalography (EEG) recordings has become a promising new tool for noninvasive epileptic focus localization. Studies to date have shown that it can be used safely and that highly localized information can be obtained. So far, no reports using comprehensive clinical information and/or long-term follow-up after epilepsy surgery in a larger patient group have been given that would allow a valuable judgment of the utility of this technique. Here, the results of 11 patients with EEG-triggered fMRI exams who also underwent presurgical evaluation of their epilepsy are given. In most patients we were able to record good quality EEG inside the magnet, allowing us to trigger fMRI acquisition by interictal discharges. The fMRI consisted of echoplanar multislice acquisition permitting a large anatomical coverage of the patient's brain. In 8 of the 11 patients the exam confirmed clinical diagnosis, either by the presence (n = 7) or absence (n = 1) of focal signal enhancement. In six patients, intracranial recordings were carried out, and in five of them, the epileptogenic zone as determined by fMRI was confirmed. Limitations were encountered a) when the focus was too close to air cavities; b) if an active epileptogenic focus was absent; and c) if only reduced cooperation with respect to body movements was provided by the patient. We conclude that EEG-triggered fMRI is a safe and powerful noninvasive tool that improves the diagnostic value of MRI by localizing the epileptic focus precisely.
