Journal: RSC Advances

Article Title: An artificial synaptic transistor using an α-In 2 Se 3 van der Waals ferroelectric channel for pattern recognition †

doi: 10.1039/d1ra07728g

Figure Lengend Snippet: Benchmarking with state of art Synapses

Article Snippet: We achieved an accuracy of 93% using grayscale MNIST (Modified National Institute of Standards and Technology) datasets.

Techniques:

Journal: RSC Advances

Article Title: An artificial synaptic transistor using an α-In 2 Se 3 van der Waals ferroelectric channel for pattern recognition †

doi: 10.1039/d1ra07728g

Figure Lengend Snippet: (a) Recognition accuracy as a function of the training cycle for different N states (10, 20, and 64). (b) Weight mapping images after 10 th and 240 th training cycles. (c) Average confusion matrix of the testing data set for 18 000 MNIST data. The digits highlighted in dark color represents the number of identified digits for all three cases, whereas those with lighter color code shows the confusion between two digits, for example digit 2, 3 for case 1. (d) Recognition accuracy as a function of the training cycle for different noise pixels introduced in the test data. Numerical digit 4 represents different % of variance introduced as background noise pixels for calculating the classification accuracy. Introducing background noise leads to difficulty in convergence state of conductance weight value.

Article Snippet: We achieved an accuracy of 93% using grayscale MNIST (Modified National Institute of Standards and Technology) datasets.

Techniques:

Journal: Scientific Reports

Article Title: Deep learning-based hemorrhage detection for diabetic retinopathy screening

doi: 10.1038/s41598-023-28680-3

Figure Lengend Snippet: Analysis of deep models using time versus classification accuracy on MNIST dataset.

Article Snippet: Image of concrete crack and Modified National Institute of Standards and Technology (MNIST) digits datasets benchmark the HemNet.

Techniques: Biomarker Discovery