Deep learning:Performance Tuning

딥러닝 최적화와 관련된 내용 정리.

Category

• im2col
• Batch Normalization
• General matrix-matrix multiplication (gemm)
• Toeplitz matrix
• Dense-Sparse-Dense Training for Deep Neural Networks (DSD)
• Deep Compression
• Pruning Neural Networks
• Dropout
• Shake-Shake regularization
• Shakeout
• Whiteout
• SGDR
• FractalNet
• [추천] Accurate Large Minibatch SGD
• Convex optimization
• [추천] Group Normalization

Caffe library

Caffe* Optimized for Intel® Architecture: Applying Modern Code Techniques

https://software.intel.com/en-us/articles/caffe-optimized-for-intel-architecture-applying-modern-code-techniques

See also

• Deep learning
• Caffe
• Caffe:Develop
• Performance Tuning
• Optuna

Favorite site

• CNNs in Practice ¹
  • [추천] CNN in practice - 고려대학교 산업경영공학과 ²
• Wikipedia (en) Toeplitz matrix - Discrete convolution (병렬 처리에 유리하다)
• [추천] 딥러닝 학습 기술들 - ratsgo's blog ³

References