اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدHighly Scalable Neuromorphic Hardware with 1-bit Stochastic nano-Synapses
46
0
0.0
(
0
)
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
Thermodynamic-driven filament formation in redox-based resistive memory and the impact of thermal fluctuations on switching probability of emerging magnetic switches are probabilistic phenomena in nature, and thus, processes of binary switching in these nonvolatile memories are stochastic and vary from switching cycle-to-switching cycle, in the same device, and from device-to-device, hence, they provide a rich in-situ spatiotemporal stochastic characteristic. This work presents a highly scalable neuromorphic hardware based on crossbar array of 1-bit resistive crosspoints as distributed stochastic synapses. The network shows a robust performance in emulating selectivity of synaptic potentials in neurons of primary visual cortex to the orientation of a visual image. The proposed model could be configured to accept a wide range of nanodevices.
قيم البحث
اقرأ أيضاً
We show that discrete synaptic weights can be efficiently used for learning in large scale neural systems, and lead to unanticipated computational performance. We focus on the representative case of learning random patterns with binary synapses in si
ngle layer networks. The standard statistical analysis shows that this problem is exponentially dominated by isolated solutions that are extremely hard to find algorithmically. Here, we introduce a novel method that allows us to find analytical evidence for the existence of subdominant and extremely dense regions of solutions. Numerical experiments confirm these findings. We also show that the dense regions are surprisingly accessible by simple learning protocols, and that these synaptic configurations are robust to perturbations and generalize better than typical solutions. These outcomes extend to synapses with multiple states and to deeper neural architectures. The large deviation measure also suggests how to design novel algorithmic schemes for optimization based on local entropy maximization.
In this work, we study the dynamic range in a neuronal network modelled by cellular automaton. We consider deterministic and non-deterministic rules to simulate electrical and chemical synapses. Chemical synapses have an intrinsic time-delay and are
susceptible to parameter variations guided by learning Hebbian rules of behaviour. Our results show that chemical synapses can abruptly enhance sensibility of the neural network, a manifestation that can become even more predominant if learning rules of evolution are applied to the chemical synapses.
Several analog and digital brain-inspired electronic systems have been recently proposed as dedicated solutions for fast simulations of spiking neural networks. While these architectures are useful for exploring the computational properties of large-
scale models of the nervous system, the challenge of building low-power compact physical artifacts that can behave intelligently in the real-world and exhibit cognitive abilities still remains open. In this paper we propose a set of neuromorphic engineering solutions to address this challenge. In particular, we review neuromorphic circuits for emulating neural and synaptic dynamics in real-time and discuss the role of biophysically realistic temporal dynamics in hardware neural processing architectures; we review the challenges of realizing spike-based plasticity mechanisms in real physical systems and present examples of analog electronic circuits that implement them; we describe the computational properties of recurrent neural networks and show how neuromorphic Winner-Take-All circuits can implement working-memory and decision-making mechanisms. We validate the neuromorphic approach proposed with experimental results obtained from our own circuits and systems, and argue how the circuits and networks presented in this work represent a useful set of components for efficiently and elegantly implementing neuromorphic cognition.
Conventional neuro-computing architectures and artificial neural networks have often been developed with no or loose connections to neuroscience. As a consequence, they have largely ignored key features of biological neural processing systems, such a
s their extremely low-power consumption features or their ability to carry out robust and efficient computation using massively parallel arrays of limited precision, highly variable, and unreliable components. Recent developments in nano-technologies are making available extremely compact and low-power, but also variable and unreliable solid-state devices that can potentially extend the offerings of availing CMOS technologies. In particular, memristors are regarded as a promising solution for modeling key features of biological synapses due to their nanoscale dimensions, their capacity to store multiple bits of information per element and the low energy required to write distinct states. In this paper, we first review the neuro- and neuromorphic-computing approaches that can best exploit the properties of memristor and-scale devices, and then propose a novel hybrid memristor-CMOS neuromorphic circuit which represents a radical departure from conventional neuro-computing approaches, as it uses memristors to directly emulate the biophysics and temporal dynamics of real synapses. We point out the differences between the use of memristors in conventional neuro-computing architectures and the hybrid memristor-CMOS circuit proposed, and argue how this circuit represents an ideal building block for implementing brain-inspired probabilistic computing paradigms that are robust to variability and fault-tolerant by design.
The neuromorphic BrainScaleS-2 ASIC comprises mixed-signal neurons and synapse circuits as well as two versatile digital microprocessors. Primarily designed to emulate spiking neural networks, the system can also operate in a vector-matrix multiplica
tion and accumulation mode for artificial neural networks. Analog multiplication is carried out in the synapse circuits, while the results are accumulated on the neurons membrane capacitors. Designed as an analog, in-memory computing device, it promises high energy efficiency. Fixed-pattern noise and trial-to-trial variations, however, require the implemented networks to cope with a certain level of perturbations. Further limitations are imposed by the digital resolution of the input values (5 bit), matrix weights (6 bit) and resulting neuron activations (8 bit). In this paper, we discuss BrainScaleS-2 as an analog inference accelerator and present calibration as well as optimization strategies, highlighting the advantages of training with hardware in the loop. Among other benchmarks, we classify the MNIST handwritten digits dataset using a two-dimensional convolution and two dense layers. We reach 98.0% test accuracy, closely matching the performance of the same network evaluated in software.
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
