Improving liquid state machines through iterative refinement of the reservoir

Norton, D. P.; Ventura, Dan

doi:10.1016/j.neucom.2010.08.005

Cited by 50 publications

(50 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Since the separability is proportional to the classification accuracy of the readout neurons [7], [15], this observation may indicate an important advantage of probabilistic reservoirs over deterministic ones.…”

Section: Resultsmentioning

confidence: 99%

“…For comparing the separation properties of different randomly generated reservoirs, we adopt an interesting metric that was recently introduced in [15]. For this procedure the responses of a reservoir to input stimuli are recorded.…”

Section: Methodsmentioning

confidence: 99%

“…Suitable liquids are commonly identified by generating numerous random networks and selecting the one that maximizes the classification accuracy of the readout function. Some interesting alternative approaches have suggested to use Hebbian learning strategies for improving the characteristics of the liquid [15], [17]. In this study, we restrict ourselves to explore a single reservoir only and leave the optimization of the liquid for future directions.…”

Section: B Setupmentioning

confidence: 99%

“…On the right hand-side, the variance occurring in the last time window is shown as a bar for each configuration. The variance at the end of the simulation is generally important for the separation of output classes and is exploited by the readout function of the LSM, as for example done in [9], [15].…”

Section: Impact Of Stochastic Parametersmentioning

confidence: 99%

See 3 more Smart Citations

Are probabilistic spiking neural networks suitable for reservoir computing?

Schliebs

Mohemmed

Kasabov

2011

The 2011 International Joint Conference on Neural Networks

View full text Add to dashboard Cite

Abstract-This study employs networks of stochastic spiking neurons as reservoirs for liquid state machines (LSM). We experimentally investigate the separation property of these reservoirs and show their ability to generalize classes of input signals. Similar to traditional LSM, probabilistic LSM (pLSM) have the separation property enabling them to distinguish between different classes of input stimuli. Furthermore, our results indicate some potential advantages of non-deterministic LSM by improving upon the separation ability of the liquid. Three non-deterministic neural models are considered and for each of them several parameter configurations are explored. We demonstrate some of the characteristics of pLSM and compare them to their deterministic counterparts. pLSM offer more flexibility due to the probabilistic parameters resulting in a better performance for some values of these parameters.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: B Setupmentioning

confidence: 99%

Section: Impact Of Stochastic Parametersmentioning

confidence: 99%

See 2 more Smart Citations

Are probabilistic spiking neural networks suitable for reservoir computing?

Schliebs

Mohemmed

Kasabov

2011

The 2011 International Joint Conference on Neural Networks

View full text Add to dashboard Cite

show abstract

“…In [81] an improved accuracy of LSM reservoir structure on pattern classification of hypothetical tasks is achieved when STDP learning was introduced into the reservoir. The learning is based on comparing the liquid states for different classes and adjusting the connection weights so that same class inputs have closer connection weights.…”

Section: Evolving Spiking Neural Networkmentioning

confidence: 99%

Evolving Spiking Neural Networks and Neurogenetic Systems for Spatio- and Spectro-Temporal Data Modelling and Pattern Recognition

Kasabov

2012

Advances in Computational Intelligence

View full text Add to dashboard Cite

Sensory Adaptation in Biomolecular Memristors Improves Reservoir Computing Performance

Maraj

Haughn

Inman

et al. 2023

Advanced Intelligent Systems

View full text Add to dashboard Cite

Despite its prevalence in neurosensory systems for pattern recognition, event detection, and learning, the effects of sensory adaptation (SA) are not explored in reservoir computing (RC). Monazomycin‐based biomolecular synapse (MzBS) devices that exhibit volatile memristance and short‐term plasticity with two strength‐dependent modes of response are studied: facilitation and facilitation‐then‐depression (i.e., SA). Their ability to perform RC tasks including digit recognition, nonlinear function learning, and aerodynamic gust classification via combination of model‐based device simulations and physical experiments where SA presence is controlled is studied. Simulations exhibiting moderate SA achieve significantly higher accuracy classifying a custom 5 × 5 binary digit set, with experimental validation achieving maximum testing accuracies of 90%. Classifications of the Modified National Institute of Standards and Technology (MNIST) handwritten digit dataset achieve a maximum testing accuracy of 94.34% in devices with SA. Fitting error of the Mackey–Glass time series is also significantly reduced by SA. Experimentally obtained pressure distributions representing gusts on an airfoil in a wind tunnel are classified by MzBS reservoirs. Reservoirs exhibiting SA achieve 100% accuracy, unlike MzBS reservoirs without SA and comparable static neural networks.

show abstract

Improving liquid state machines through iterative refinement of the reservoir

Cited by 50 publications

References 17 publications

Are probabilistic spiking neural networks suitable for reservoir computing?

Are probabilistic spiking neural networks suitable for reservoir computing?

Evolving Spiking Neural Networks and Neurogenetic Systems for Spatio- and Spectro-Temporal Data Modelling and Pattern Recognition

Sensory Adaptation in Biomolecular Memristors Improves Reservoir Computing Performance

Contact Info

Product

Resources

About