Porovnat metody
Prohlédněte si vybrané metody vedle sebe; řádky, které se liší, jsou zvýrazněny.
| LoRA a PEFT× | Generativní adversariální síť× | Vision Transformer× | |
|---|---|---|---|
| Obor | Hluboké učení | Hluboké učení | Hluboké učení |
| Rodina | Machine learning | Machine learning | Machine learning |
| Rok vzniku≠ | 2022 | 2014 | 2021 |
| Tvůrce≠ | Hu, E. J. et al.; Lester, B. et al. | Goodfellow, I. et al. | Dosovitskiy, A. et al. |
| Typ≠ | Parameter-efficient fine-tuning of large pretrained models | Generative deep learning (adversarial two-network game) | Transformer architecture for images (self-attention over patches) |
| Původní zdroj≠ | Hu, E. J. et al. (2022). LoRA: Low-Rank Adaptation of Large Language Models. ICLR. link ↗ | Goodfellow, I. et al. (2014). Generative Adversarial Nets. NeurIPS. link ↗ | Dosovitskiy, A. et al. (2021). An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale. ICLR. link ↗ |
| Další názvy≠ | LoRA ve PEFT — Parametre Verimli İnce Ayar, Low-Rank Adaptation, parameter-efficient fine-tuning, prefix tuning | Üretici Çekişmeli Ağ (GAN), GAN, generative adversarial nets, adversarial network | Görsel Transformer (ViT), görsel transformer, ViT, patch transformer for images |
| Příbuzné≠ | 5 | 4 | 5 |
| Shrnutí≠ | LoRA (Low-Rank Adaptation), introduced by Hu et al. in 2022, and the broader family of parameter-efficient fine-tuning (PEFT) methods adapt large pretrained language models to new tasks by training only a small number of extra parameters instead of every weight in the model. This makes fine-tuning possible with far less GPU memory and compute while leaving the original model largely untouched. | A Generative Adversarial Network (GAN), introduced by Ian Goodfellow and colleagues in 2014, produces realistic synthetic data through the competition of two neural networks — a generator and a discriminator. It is widely used for image synthesis, data augmentation, and distribution estimation. | The Vision Transformer (ViT), introduced by Dosovitskiy and colleagues in 2021, splits an image into fixed-size patches, treats those patches as a sequence, and applies the Transformer self-attention mechanism to image classification. Given enough training data, it surpasses convolutional neural networks (CNNs). |
| ScholarGateDatová sada ↗ |
|
|
|