Question
These neural networks may include pooling layers to perform dimension reduction. For 10 points each:
[10h] Name these neural networks common in image identification that perform feature extraction via layers that apply a namesake kernel function to square patches of the input.
ANSWER: convolutional neural network [or CNNs]
[10e] When feeding images into a convolutional neural network, the channels of each of these elements must be separated. Computer images consist of these tiny squares.
ANSWER: pixels
[10m] To assess performance and avoid overfitting, it is typical to split input data into test data and this type of data. This type of data is labeled in supervised machine learning and is subdivided into validation sets in cross-validation.
ANSWER: training data set [or training set]
<Science - Other Science>
Summary
| 2023 ACF Fall (Rutgers) | 10/14/2023 | Y | 4 | 20.00 | 100% | 50% | 50% |
Data
| Maryland | Fordham | 10 | 10 | 0 | 20 |
| Rowan | Lehigh | 0 | 10 | 10 | 20 |
| Penn | Columbia B | 0 | 10 | 0 | 10 |
| Columbia A | Princeton | 10 | 10 | 10 | 30 |