MobileNet MCQ 15 Questions
Time: ~25 mins Advanced

MobileNet MCQ

Depthwise separable convolutions slash compute so models run on phones and embedded devices.

Easy: 5 Q Medium: 6 Q Hard: 4 Q
Mobile

Edge deploy

Depthwise

Per-channel

Pointwise

1×1 mix

Multipliers

α, ρ

Efficient CNNs

MobileNet factorizes a standard convolution into a depthwise spatial filter per input channel followed by a pointwise 1×1 that mixes channels. This cuts FLOPs and parameters dramatically. Width and input resolution multipliers offer accuracy–latency tradeoffs.

Depthwise separable

Cost roughly channels × k² + channels² vs channels_in × channels_out × k² for a k×k conv.

Key ideas

Depthwise

Each channel convolved independently with its own spatial kernel.

Pointwise

1×1 conv combines channels—linear mix at each pixel.

Width multiplier α

Uniformly thins channel counts across the network.

Resolution

Smaller input ρ reduces compute quadratically in spatial size.

MobileNet block

depthwise 3×3 → BN → ReLU → pointwise 1×1 → BN → ReLU

Pro tip: MobileNetV2 adds inverted residuals and linear bottlenecks; still built on depthwise separable ideas.