Mixed Precision

---

📌 1. Mixed Precision Training

• Mixed Precision Training
  • NVIDIA 공식 논문:
    https://arxiv.org/abs/1710.03740
  • 관련 공식 가이드:
    https://docs.nvidia.com/deeplearning/performance/mixed-precision-training/index.html

---

📌 2. EfficientNet (경량화 모델)

• EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks
  • https://arxiv.org/abs/1905.11946
• EfficientNetV2: Smaller Models and Faster Training
  • https://arxiv.org/abs/2104.00298

---

📌 3. Dropout & Regularization

• Dropout: A Simple Way to Prevent Neural Networks from Overfitting
  • https://jmlr.org/papers/v15/srivastava14a.html
  • PDF: http://www.cs.toronto.edu/~hinton/absps/JMLRdropout.pdf
• Understanding the Disharmony between Dropout and Batch Normalization by Variance Shift
  • https://arxiv.org/abs/1801.05134

---

📌 4. Data Augmentation Techniques

• A survey on Image Data Augmentation for Deep Learning
  • https://arxiv.org/abs/1712.04621

---

📌 5. Model Compression & Quantization

• Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference
  • https://arxiv.org/abs/1712.05877
• Post-training Quantization for TensorFlow Lite
  • 공식 가이드:
    https://www.tensorflow.org/lite/performance/post_training_quantization

---

📌 6. TensorFlow Dataset Performance

• Performance guide for tf.data
  • https://www.tensorflow.org/guide/data_performance

---

📌 7. General Deep Learning Optimization

• Bag of Tricks for Image Classification with Convolutional Neural Networks
  • https://arxiv.org/abs/1812.01187

---

✅ 추천 논문/자료 요약

주제	링크
Mixed Precision	NVIDIA 논문
EfficientNet	EfficientNet 논문
Dropout	Dropout 원 논문
Data Augmentation	Survey 논문
Quantization	Quantization 논문
Dataset 최적화	TensorFlow guide

---