Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userRandom telegraph signal analysis with a recurrent neural network
327
0
0.0
(
0
)
Ask ChatGPT about the research
No Arabic abstract
We use an artificial neural network to analyze asymmetric noisy random telegraph signals (RTSs), and extract underlying transition rates. We demonstrate that a long short-term memory neural network can vastly outperform conventional methods, particularly for noisy signals. Our technique gives reliable results as the signal-to-noise ratio approaches one, and over a wide range of underlying transition rates. We apply our method to random telegraph signals generated by a superconducting double dot based photon detector, allowing us to extend our measurement of quasiparticle dynamics to new temperature regimes.
rate research
Read More
A general method is presented to explicitly compute autocovariance functions for non-Poisson dichotomous noise based on renewal theory. The method is specialized to a random telegraph signal of Mittag-Leffler type. Analytical predictions are compared to Monte Carlo simulations. Non-Poisson dichotomous noise is non-stationary and standard spectral methods fail to describe it properly as they assume stationarity.
We consider the problem of sparse signal recovery from a small number of random projections (measurements). This is a well known NP-hard to solve combinatorial optimization problem. A frequently used approach is based on greedy iterative procedures, such as the Matching Pursuit (MP) algorithm. Here, we discuss a fast GPU implementation of the MP algorithm, based on the recently released NVIDIA CUDA API and CUBLAS library. The results show that the GPU version is substantially faster (up to 31 times) than the highly optimized CPU version based on CBLAS (GNU Scientific Library).
Head motion is inevitable in the acquisition of diffusion-weighted images, especially for certain motion-prone subjects and for data gathering of advanced diffusion models with prolonged scan times. Deficient accuracy of motion correction cause deterioration in the quality of diffusion model reconstruction, thus affecting the derived measures. This results in either loss of data, or introducing bias in outcomes from data of different motion levels, or both. Hence minimizing motion effects and reutilizing motion-contaminated data becomes vital to quantitative studies. We have previously developed a 3-dimensional hierarchical convolution neural network (3D H-CNN) for robust diffusion kurtosis mapping from under-sampled data. In this study, we propose to extend this method to motion-contaminated data for robust recovery of diffusion model-derived measures with a process of motion assessment and corrupted volume rejection. We validate the proposed pipeline in two in-vivo datasets. Results from the first dataset of individual subjects show that all the diffusion tensor and kurtosis tensor-derived measures from the new pipeline are minimally sensitive to motion effects, and are comparable to the motion-free reference with as few as eight volumes retained from the motion-contaminated data. Results from the second dataset of a group of children with attention deficit hyperactivity disorder demonstrate the ability of our approach in ameliorating spurious group differences due to head motion. This method shows great potential for exploiting some valuable but motion-corrupted DWI data which are likely to be discarded otherwise, and applying to data with different motion level thus improving their utilization and statistic power.
We present a new Monte Carlo Markov Chain algorithm for CMB analysis in the low signal-to-noise regime. This method builds on and complements the previously described CMB Gibbs sampler, and effectively solves the low signal-to-noise inefficiency problem of the direct Gibbs sampler. The new algorithm is a simple Metropolis-Hastings sampler with a general proposal rule for the power spectrum, C_l, followed by a particular deterministic rescaling operation of the sky signal. The acceptance probability for this joint move depends on the sky map only through the difference of chi-squared between the original and proposed sky sample, which is close to unity in the low signal-to-noise regime. The algorithm is completed by alternating this move with a standard Gibbs move. Together, these two proposals constitute a computationally efficient algorithm for mapping out the full joint CMB posterior, both in the high and low signal-to-noise regimes.
Digital hologram rendering can be performed by a convolutional neural network, trained with image pairs calculated by numerical wave propagation from sparse generating images. 512-by-512 pixeldigital Gabor magnitude holograms are successfully estimated from experimental interferograms by a standard UNet trained with 50,000 synthetic image pairs over 70 epochs.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
