> ML_LITERATURE // SANDLER-2018-MOBILENETV2-INVERTED-RESIDUALS-LINEAR-BOTTLENECKS_v1.0
MobileNetV2: Inverted Residuals and Linear Bottlenecks
Mark Sandler, Andrew Howard, Menglong Zhu, Andrey Zhmoginov, Liang-Chieh Chen · IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2018)
seminal-architecture2018foundationalthirdPartyReproduced
Principal Contribution
Introduced inverted residual blocks with linear bottlenecks, expanding to high dimensions before depthwise filtering and projecting back to preserve non-linear manifold capacity.
Operational Relevance
Serves as qualified reference for deploying task-image-classification, task-object-detection in production.
Assumptions
- Spatial feature coherence and data manifold structure adhere to continuous representation hypotheses
Limitations
- Computational complexity scales with spatial resolution and parameter capacity
Connected Algorithms, Architectures & Tools
Related Algorithms:
Related Architectures:
Implementing Libraries: