Implements a densely connected neural network architecture to augment password security through biometric keystroke analysis. When tested on the CMU Keystroke Dynamics Benchmark Dataset, the model reports a 0% false positive rate and a sub 1% false negative rate.
Model: "sequential_1"
_________________________________________________________________
Layer (type) Output Shape Param #
=================================================================
dense_1 (Dense) (None, 256) 8192
_________________________________________________________________
dropout_1 (Dropout) (None, 256) 0
_________________________________________________________________
dense_2 (Dense) (None, 128) 32896
_________________________________________________________________
batch_normalization_1 (Batch (None, 128) 512
_________________________________________________________________
dense_3 (Dense) (None, 128) 16512
_________________________________________________________________
dropout_2 (Dropout) (None, 128) 0
_________________________________________________________________
dense_4 (Dense) (None, 64) 8256
_________________________________________________________________
batch_normalization_2 (Batch (None, 64) 256
_________________________________________________________________
dense_5 (Dense) (None, 32) 2080
_________________________________________________________________
dropout_3 (Dropout) (None, 32) 0
_________________________________________________________________
dense_6 (Dense) (None, 16) 528
_________________________________________________________________
batch_normalization_3 (Batch (None, 16) 64
_________________________________________________________________
dense_7 (Dense) (None, 2) 34
=================================================================
Total params: 69,330
Trainable params: 68,914
Non-trainable params: 416
_________________________________________________________________