Question

The components of these systems can have their characteristics updated according to the Kohonen rule. François Chollet developed the Keras library to interface with these systems. One type of these systems is characterized by the (15[2])hyperparameters of depth, stride, and padding, and contains dropout and pooling components. That type of these systems involves smoothing adjacent pixels by (*) convolving them together. The gradient of the loss function is updated in a dynamic (10[1])programming algorithm for these systems called backpropagation. ReLu is an example of a nonlinear activation function used by the nodes of these systems, which are called “deep” if they have multiple layers. (10[1])For 10 points, identify these machine learning (10[1])constructs named for mimicking the human (10[1])brain. ■END■

ANSWER: artificial neural networks [or neural nets or ANNs; accept convolutional neural networks or CNNs; accept deep neural networks; prompt on deep learning models or machine learning models or ML models or equivalents]
<Science - Other Science>
= Average correct buzz position

Back to tossups

Buzzes

PlayerTeamOpponentBuzz PositionValue
Matthew SiffYaleBard3515
Andrew YangColumbia CColumbia A3515
Eshan PantNYU ANYU B7110
Lexi TermanRutgersNYU C10310
Miles JaffeeLehigh ALehigh B11010
Jacob DubnerColumbia BGatherer11610

Summary

2024 Booster Shot (Columbia)02/23/2024Y6100%33%0%78.33
2024 Booster Shot (Waterloo)02/23/2024Y4100%50%50%66.25
2024 Booster Shot (Vanderbilt)03/02/2024Y4100%50%0%68.25
2024 Booster Shot (Great Lakes)03/09/2024Y6100%33%17%69.17
2024 Booster Shot (Great Lakes)03/09/2024Y1100%100%0%68.00
2024 Booster Shot (WUSTL)03/09/2024Y3100%67%0%56.67