Dataset Information

Time and rate dependent synaptic learning in neuro-mimicking resistive memories.

ABSTRACT: Memristors have demonstrated immense potential as building blocks in future adaptive neuromorphic architectures. Recently, there has been focus on emulating specific synaptic functions of the mammalian nervous system by either tailoring the functional oxides or engineering the external programming hardware. However, high device-to-device variability in memristors induced by the electroforming process and complicated programming hardware are among the key challenges that hinder achieving biomimetic neuromorphic networks. Here, a simple hybrid complementary metal oxide semiconductor (CMOS)-memristor approach is reported to implement different synaptic learning rules by utilizing a CMOS-compatible memristor based on oxygen-deficient SrTiO_3-x (STO_x). The potential of such hybrid CMOS-memristor approach is demonstrated by successfully imitating time-dependent (pair and triplet spike-time-dependent-plasticity) and rate-dependent (Bienenstosk-Cooper-Munro) synaptic learning rules. Experimental results are benchmarked against in-vitro measurements from hippocampal and visual cortices with good agreement. The scalability of synaptic devices and their programming through a CMOS drive circuitry elaborates the potential of such an approach in realizing adaptive neuromorphic computation and networks.

SUBMITTER: Ahmed T

PROVIDER: S-EPMC6817848 | biostudies-literature | 2019 Oct

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Time and rate dependent synaptic learning in neuro-mimicking resistive memories.

Ahmed Taimur T Walia Sumeet S Mayes Edwin L H ELH Ramanathan Rajesh R Bansal Vipul V Bhaskaran Madhu M Sriram Sharath S Kavehei Omid O

Scientific reports 20191028 1

Memristors have demonstrated immense potential as building blocks in future adaptive neuromorphic architectures. Recently, there has been focus on emulating specific synaptic functions of the mammalian nervous system by either tailoring the functional oxides or engineering the external programming hardware. However, high device-to-device variability in memristors induced by the electroforming process and complicated programming hardware are among the key challenges that hinder achieving biomimet ...[more]

PMID: 31659247

Similar Datasets

Project description:UnlabelledAmong the most fundamental questions in viral evolutionary biology are how fast viruses evolve and how evolutionary rates differ among viruses and fluctuate through time. Traditionally, viruses are loosely classed into two groups: slow-evolving DNA viruses and fast-evolving RNA viruses. As viral evolutionary rate estimates become more available, it appears that the rates are negatively correlated with the measurement timescales and that the boundary between the rates of DNA and RNA viruses might not be as clear as previously thought. In this study, we collected 396 viral evolutionary rate estimates across almost all viral genome types and replication strategies, and we examined their rate dynamics. We showed that the time-dependent rate phenomenon exists across multiple levels of viral taxonomy, from the Baltimore classification viral groups to genera. We also showed that, by taking the rate decay dynamics into account, a clear division between the rates of DNA and RNA viruses as well as reverse-transcribing viruses could be recovered. Surprisingly, despite large differences in their biology, our analyses suggested that the rate decay speed is independent of viral types and thus might be useful for better estimation of the evolutionary time scale of any virus. To illustrate this, we used our model to reestimate the evolutionary timescales of extant lentiviruses, which were previously suggested to be very young by standard phylogenetic analyses. Our analyses suggested that these viruses are millions of years old, in agreement with paleovirological evidence, and therefore, for the first time, reconciled molecular analyses of ancient and extant viruses.ImportanceThis work provides direct evidence that viral evolutionary rate estimates decay with their measurement timescales and that the rate decay speeds do not differ significantly among viruses despite the vast differences in their molecular features. After adjustment for the rate decay dynamics, the division between the rates of double-stranded DNA (dsDNA), single-stranded RNA (ssRNA), and ssDNA/reverse-transcribing viruses could be seen more clearly than before. Our results provide a guideline for further improvement of the molecular clock. As a demonstration of this, we used our model to reestimate the timescales of modern lentiviruses, which were previously thought to be very young, and concluded that they are millions of years old. This result matches the estimate from paleovirological analyses, thus bridging the gap between ancient and extant viral evolutionary studies.

Dataset Information

Time and rate dependent synaptic learning in neuro-mimicking resistive memories.

Publications

Time and rate dependent synaptic learning in neuro-mimicking resistive memories.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets