Abstract
Noise exists in nearly all physical systems ranging from simple electronic devices such as transistors to complex systems such as neural networks. To understand a system's behavior, it is vital to know the origin of the noise and its characteristics. Recently, it was shown that the nonlinear electronic properties of a disordered dopant atom network in silicon can be exploited for efficiently executing classification tasks through “material learning.” Here, we study the dopant network's intrinsic 1/f noise arising from Coulomb interactions, and its impact on the features that determine its computational abilities, viz., the nonlinearity and the signal‐to‐noise ratio (SNR), is investigated. The findings on optimal SNR and nonlinear transformation of data by this nonlinear network provide a guideline for the scaling of physical learning machines and shed light on neuroscience from a new perspective.
Highlights
Noise exists in most physical systems ranging from simple electronic devices In doped semiconductors, the 1/f noise such as transistors to complex systems such as neural networks
Intrinsic 1/f noise arising from Coulomb interactions, and its impact on the hypothetically because the large-scale features that determine its computational abilities, viz., the nonlinearity and the signal-to-noise ratio (SNR), is investigated
The findings on optimal SNR and nonlinear transformation of data by this nonlinear network provide a guideline for the scaling of physical learning machines and shed light on neucomplex neural networks are poised at criticality,[6,10,11,12] i.e., at the border of a phase transition such as the onset of synchronous activity.[4,12]
Summary
To investigate how the DNPU responds to external stimulation, we applied a small sinusoidal voltage signal (0.1 V amplitude, 1 Hz frequency, see Experimental Section and Supporting Information) to the gate electrode (Figure 1B, inset) and recorded the DC for different bias voltages applied to the source.[14] The gate electrode is far away from the source and drain electrodes. A 1 Hz signal emerges when the bias voltage crosses the threshold VSD,th (Figure 3, upper inset). The SNR maximizes at a certain noise intensity, a mechanism known as noise-induced threshold crossing.[29] Our present study, implies that the DNPU’s response to an external signal maximizes when the system is energized (voltage biased) optimally with respect to its internal (1/f ) noise. The bias voltage should not be too large, to not overshadow the effects of the external stimulation
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call