> ML_LITERATURE // TAN-2019-EFFICIENTNET-RETHINKING-MODEL-SCALING-CNNS_v1.0
EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks
Mingxing Tan, Quoc V. Le · International Conference on Machine Learning (ICML) (2019)
seminal-architecture2019industry-standardthirdPartyReproduced
Principal Contribution
Invented compound scaling, systematically balancing network depth, width, and image resolution with a fixed compound coefficient.
Operational Relevance
Serves as qualified reference for implementing task-image-classification, task-feature-extraction in production systems.
Assumptions
- Underlying spatio-temporal continuity and domain distributional stability hold
Limitations
- Performance scaling and computational footprint depend on receptive field depth, sequence length, and resolution
Connected Algorithms, Architectures & Tools
Related Algorithms:
Related Architectures:
Implementing Libraries: