Which of the following are true about Depth wise-separable convolutions? (Choose all that apply)

Depthwise separable convolutions are a type of convolutional operation used in deep learning, particularly in lightweight models like MobileNets. Here are some true statements about them:

1. Reduction of Parameters: Depthwise separable convolutions significantly reduce the number of parameters compared to traditional convolutions by using a depthwise convolution followed by a pointwise convolution.

2. Separate Spatial and Channel-wise Operations: They perform spatial convolution independently for each channel (depthwise convolution) followed by a 1x1 convolution across all channels (pointwise convolution).

3. Efficient Computation: They are computationally more efficient than standard convolutions, which helps in reducing the model size and improving inference speed.

4. Maintaining Performance: Despite the reduction in parameters and computations, they can achieve comparable performance to standard convolutions, particularly in image classification tasks.

5. Used in Mobile Networks: Depthwise separable convolutions are commonly used in architectures designed for mobile devices due to their efficiency.

Please choose the relevant true statements based on the options provided.