Auto Topic: cnns
auto_cnns | topic
Coverage Score
1
Mentioned Chunks
7
Mentioned Docs
1
Required Dimensions
definitionpros_cons
Covered Dimensions
definitionpros_cons
Keywords
cnns
Relations
| Source | Type | Target | W |
|---|---|---|---|
| Auto Topic: cnns | CO_OCCURS | Auto Topic: convolutional | 4 |
| Auto Topic: cnns | CO_OCCURS | Auto Topic: dimension | 4 |
| Auto Topic: cnns | CO_OCCURS | Auto Topic: convolution | 3 |
| Auto Topic: cnns | CO_OCCURS | Auto Topic: kernel | 3 |
| Auto Topic: cnns | CO_OCCURS | Auto Topic: stride | 3 |
| Auto Topic: cnn | CO_OCCURS | Auto Topic: cnns | 3 |
Evidence Chunks
| Source | Confidence | Mentions | Snippet |
|---|---|---|---|
textbook Artificial-Intelligence-A-Modern-Approach-4th-Edition.pdf | 0.59 | 3 | ... layer size. It’s also possible to reduce the layer size simply by having a fully connected layer with fewer units than the preceding layer. CNNs often have one or two such layers preceding the final softmax layer. 22.3.2 Tensor operations in CNNs We saw in Equations (22.1) and (22 ... |
textbook Artificial-Intelligence-A-Modern-Approach-4th-Edition.pdf | 0.57 | 2 | ... -dimensional special cases of tensors, which (in deep learning terminology) are simply multidimensional arraysTensor of any dimension.5 For CNNs, tensors are a way of keeping track of the “shape” of the data as it progresses through the layers of the network. This is important be ... |
textbook Artificial-Intelligence-A-Modern-Approach-4th-Edition.pdf | 0.57 | 2 | e effect. 27.4.1 Image classification with convolutional neural networks Convolutional neural networks (CNNs)are spectacularly successful image classifiers. With enough training data and enough training ingenuity, CNNs produce very successful classifi- cation systems, much better th ... |
textbook Artificial-Intelligence-A-Modern-Approach-4th-Edition.pdf | 0.55 | 1 | ... the fact that convolution is a linear matrix operation serves as a reminder that gradient descent can be applied easily and effectively to CNNs, just as it can to plain vanilla neural networks. As mentioned earlier, there will be d kernels, not just one; so, with a stride of 1, t ... |
textbook Artificial-Intelligence-A-Modern-Approach-4th-Edition.pdf | 0.55 | 1 | ... however, this additional “kernel dimension” does not have any adjacency properties, so it does not make sense to run convolutions along it. CNNs were inspired originally by models of the visual cortex proposed in neuroscience. In those models, the receptive field of a neuron is th ... |
textbook Artificial-Intelligence-A-Modern-Approach-4th-Edition.pdf | 0.55 | 1 | ... s, and computing resources, the top-5 error rate has been reduced to less than 2%—well below the error rate of a trained human (around 5%). CNNs have been applied to a wide range of vision tasks, from self-driving cars to grading cucumbers. 8 Driving, which is covered in Section ... |
textbook Artificial-Intelligence-A-Modern-Approach-4th-Edition.pdf | 0.55 | 1 | ... in training, and so can adjust for the effect. 27.4.1 Image classification with convolutional neural networks Convolutional neural networks (CNNs)are spectacularly successful image classifiers. With enough training data and enough training |