A binary classifier based on a reconfigurable dense network of metallic nanojunctions

Major efforts to reproduce the brain performances in terms of classification and pattern recognition have been focussed on the development of artificial neuromorphic systems based on top-down lithographic technologies typical of highly integrated components of digital computers. Unconventional computing has been proposed as an alternative exploiting the complexity and collective phenomena originating from various classes of physical substrates. Materials composed of a large number of non-linear nanoscale junctions are of particular interest: these systems, obtained by the self-assembling of nano-objects like nanoparticles and nanowires, results in non-linear conduction properties characterized by spatiotemporal correlation in their electrical activity. This appears particularly useful for classification of complex features: nonlinear projection into a high-dimensional space can make data linearly separable, providing classification solutions that are computationally very expensive with digital computers. Recently we reported that nanostructured Au films fabricated from the assembling of gold clusters by supersonic cluster beam deposition show a complex resistive switching behaviour. Their non-linear electric behaviour is remarkably stable and reproducible allowing the facile training of the devices on precise resistive states. Here we report about the fabrication and characterization of a device that allows the binary classification of Boolean functions by exploiting the properties of cluster-assembled Au films interconnecting a generic pattern of electrodes. This device, that constitutes a generalization of the perceptron, can receive inputs from different electrode configurations and generate a complete set of Boolean functions of n variables for classification tasks. We also show that the non-linear and non-local electrical conduction of cluster-assembled gold films, working at room temperature, allows the classification of non-linearly separable functions without previous training of the device.

[1]  Hermann Kohlstedt,et al.  A memristive plasticity model of voltage-based STDP suitable for recurrent bidirectional neural networks in the hippocampus , 2018, Scientific Reports.

[2]  Audrius V. Avizienis,et al.  Emergent Criticality in Complex Turing B‐Type Atomic Switch Networks , 2012, Advanced materials.

[3]  Marvin Minsky,et al.  Perceptrons: An Introduction to Computational Geometry , 1969 .

[4]  Yusuf Leblebici,et al.  Neuromorphic computing with multi-memristive synapses , 2017, Nature Communications.

[5]  D. R. Strachan,et al.  Real-time TEM imaging of the formation of crystalline nanoscale gaps. , 2008, Physical review letters.

[6]  Susan Stepney,et al.  A substrate-independent framework to characterize reservoir computers , 2018, Proceedings of the Royal Society A.

[7]  Celestine Preetham Lawrence,et al.  Evolution of a designless nanoparticle network into reconfigurable Boolean logic. , 2015, Nature nanotechnology.

[8]  P. Milani,et al.  Anomalous electrical conduction and negative temperature coefficient of resistance in nanostructured gold resistive switching films , 2020, Scientific Reports.

[9]  P. Buffat,et al.  Size effect on the melting temperature of gold particles , 1976 .

[10]  Saurabh K. Bose,et al.  Stable Self-Assembled Atomic-Switch Networks for Neuromorphic Applications , 2017, IEEE Transactions on Electron Devices.

[11]  Yoshua Bengio,et al.  Random Search for Hyper-Parameter Optimization , 2012, J. Mach. Learn. Res..

[12]  Christof Teuscher,et al.  Online Learning in a Chemical Perceptron , 2013, Artificial Life.

[13]  H. Jaeger,et al.  Unconventional Information Processing Systems , Novel Hardware : A Tour d ’ Horizon , 2017 .

[14]  Iijima,et al.  Structural instability of ultrafine particles of metals. , 1986, Physical review letters.

[15]  An-Ping Li,et al.  Large discrete resistance jump at grain boundary in copper nanowire. , 2010, Nano letters.

[16]  D. Gall.  Electron mean free path in elemental metals , 2016 .

[17]  Benjamin J Wiley,et al.  Emergence of winner-takes-all connectivity paths in random nanowire networks , 2018, Nature Communications.

[18]  Wei Yang Lu,et al.  Nanoscale memristor device as synapse in neuromorphic systems. , 2010, Nano letters.

[19]  Laurent Larger,et al.  High-Speed Photonic Reservoir Computing Using a Time-Delay-Based Architecture: Million Words per Second Classification , 2017 .

[20]  D. R. Strachan,et al.  Memristive switching of single-component metallic nanowires , 2010, Nanotechnology.

[21]  Fabio L. Traversa,et al.  Memcomputing: Leveraging memory and physics to compute efficiently , 2018, ArXiv.

[22]  Fabien Alibart,et al.  Pattern classification by memristive crossbar circuits using ex situ and in situ training , 2013, Nature Communications.

[23]  Fabio Benfenati,et al.  Brain‐Inspired Structural Plasticity through Reweighting and Rewiring in Multi‐Terminal Self‐Organizing Memristive Nanowire Networks , 2020, Adv. Intell. Syst..

[24]  Andrea Bellacicca,et al.  Facile fabrication of complex networks of memristive devices , 2017, Scientific Reports.

[25]  M. Welland,et al.  Size effects in the electrical resistivity of polycrystalline nanowires , 2000 .

[26]  Masakazu Aono,et al.  Atomic switch networks as complex adaptive systems , 2018 .

[27]  Adam Z. Stieg,et al.  Emergent dynamics of neuromorphic nanowire networks , 2019, Scientific Reports.

[28]  Tao Chen,et al.  Classification with a disordered dopant-atom network in silicon , 2020, Nature.

[29]  P. Milani,et al.  Non-ohmic behavior and resistive switching of Au cluster-assembled films beyond the percolation threshold , 2019, Nanoscale Advances.

[30]  Paolo Milani,et al.  The influence of nanoscale morphology on the resistivity of cluster-assembled nanostructured metallic thin films , 2010 .

[31]  Eric J. Sandouk,et al.  Multistate resistive switching in silver nanoparticle films , 2015, Science and technology of advanced materials.

[32]  Paolo Milani,et al.  Cluster beam deposition: a tool for nanoscale science and technology , 2006 .

[33]  P. Milani,et al.  Production and characterization of highly intense and collimated cluster beams by inertial focusing in supersonic expansions , 2001 .

[34]  V. Saxena,et al.  Reconfigurable Threshold Logic Gates using memristive devices , 2012, 2012 IEEE Subthreshold Microelectronics Conference (SubVT).

[35]  Ali Khiat,et al.  Seamlessly fused digital-analogue reconfigurable computing using memristors , 2018, Nature Communications.

[36]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[37]  Masakazu Aono,et al.  A theoretical and experimental study of neuromorphic atomic switch networks for reservoir computing , 2013, Nanotechnology.

[38]  Christof Teuscher.  Unconventional Computing Catechism , 2014, Front. Robot. AI.

[39]  A. Halbritter,et al.  Connective neck evolution and conductance steps in hot point contacts , 2001, cond-mat/0112167.

[40]  Mark E. Welland,et al.  Analysis of failure mechanisms in electrically stressed Au nanowires , 1999 .

[41]  Julian Francis Miller,et al.  Evolution-in-materio: evolving computation in materials , 2014, Evolutionary Intelligence.

[42]  P. Milani,et al.  Complex electrical spiking activity in resistive switching nanostructured Au two-terminal devices , 2020, Nanotechnology.

[43]  Paolo Milani,et al.  RAPID COMMUNICATION: A pulsed microplasma source of high intensity supersonic carbon cluster beams , 1999 .

[44]  John W. Backus,et al.  Can programming be liberated from the von Neumann style?: a functional style and its algebra of programs , 1978, CACM.

[45]  Geoffrey E. Hinton,et al.  Learning representations by back-propagating errors , 1986, Nature.

[46]  Russell A. Poldrack,et al.  Principles of dynamic network reconfiguration across diverse brain states , 2017, NeuroImage.

[47]  G. Edelman,et al.  Complexity and coherency: integrating information in the brain , 1998, Trends in Cognitive Sciences.

[48]  Herbert Jaeger,et al.  Towards a generalized theory comprising digital, neuromorphic and unconventional computing , 2021, Neuromorph. Comput. Eng..

[49]  Doo Seok Jeong,et al.  Towards artificial neurons and synapses: a materials point of view , 2013 .

[50]  Pritish Narayanan,et al.  Equivalent-accuracy accelerated neural-network training using analogue memory , 2018, Nature.

[51]  R. Munoz,et al.  Size effects and charge transport in metals: Quantum theory of the resistivity of nanometric metallic structures arising from electron scattering by grain boundaries and by rough surfaces , 2017 .

[52]  G W Hoffmann,et al.  A neural network model based on the analogy with the immune system. , 1986, Journal of theoretical biology.

[53]  Shree Krishna Acharya,et al.  Avalanches and criticality in self-organized nanoscale networks , 2019, Science Advances.

[54]  Chiara Bartolozzi,et al.  Neuromorphic Electronic Circuits for Building Autonomous Cognitive Systems , 2014, Proceedings of the IEEE.

[55]  J. W. Backus,et al.  Can programming be liberated from the von Neumann style , 1977 .

[56]  Eduardo Perez,et al.  Neuromorphic on-chip recognition of saliva samples of COPD and healthy controls using memristive devices , 2020, Scientific Reports.

[57]  Julian Francis Miller,et al.  Evolution-in-materio: solving computational problems using carbon nanotube–polymer composites , 2015, Soft Computing.

[58]  Damien Querlioz,et al.  Using Memristors for Robust Local Learning of Hardware Restricted Boltzmann Machines , 2019, Scientific Reports.

[59]  S. Schultz.  Principles of Neural Science, 4th ed. , 2001 .

[60]  J. B. Mallinson,et al.  Synaptic dynamics in complex self-assembled nanoparticle networks. , 2018, Faraday discussions.

[61]  Julian Francis Miller,et al.  The alchemy of computation: designing with the unknown , 2019, Natural Computing.

[62]  Toshiyuki Yamane,et al.  Recent Advances in Physical Reservoir Computing: A Review , 2018, Neural Networks.

[63]  P. Milani,et al.  Electrical conduction in nanogranular cluster-assembled metallic films , 2021, Advances in Physics: X.

[64]  Giacomo Indiveri,et al.  A differential memristive synapse circuit for on-line learning in neuromorphic computing systems , 2017, ArXiv.

[65]  N. Otsu.  A threshold selection method from gray level histograms , 1979 .

[66]  Giacomo Indiveri,et al.  Memory and Information Processing in Neuromorphic Systems , 2015, Proceedings of the IEEE.