文章目录
Latent space
Source
Understanding Latent Space in Machine Learning | by Ekin Tiu | Towards Data Science
Key Takeaways
- The latent space is simply a representation of compressed data in which similar data points are closer together in space.
- Latent space is useful for learning data features and for finding simpler representations of data for analysis.
- We can understand patterns or structural similarities between data points by analyzing data in the latent space, be it through manifolds, clustering, etc.
- We can interpolate data in the latent space, and use our model’s decoder to ‘generate’ data samples.
- We can visualize the latent space using algorithms such as t-SNE and LLE, which takes our latent space representation and transforms it into 2D or 3D.
- 隐空间只是一种压缩数据的表示，其中相似的数据点在空间上更接近。
- 隐空间对于学习数据特征、寻找更简单的数据表征从而进行分析很有帮助。
- 我们可以通过分析隐空间中的数据来了解数据点之间的模式或结构相似性，无论是通过流形（manifolds）还是聚类等方式。
- 我们可以在隐空间中插值数据，并使用我们模型的解码器来“生成”数据样本。
- 我们可以使用t-SNE和LLE等算法将隐空间可视化，这些算法将我们的隐空间表征转化为二维或三维。
18 Impressive Applications of GANs
Source
18 Impressive Applications of Generative Adversarial Networks (GANs) (machinelearningmastery.com)
Key Takeaways
- Generate Examples for Image Datasets
- Generate Photographs of Human Faces
- Generate Realistic Photographs
- Generate Cartoon Characters
- Image-to-Image Translation
- Text-to-Image Translation
- Semantic-Image-to-Photo Translation
- Face Frontal View Generation
- Generate New Human Poses
- Photos to Emojis
- Photograph Editing
- Face Aging
- Photo Blending
- Super Resolution
- Photo Inpainting
- Clothing Translation
- Video Prediction
- 3D Object Generation
Explore the GAN Latent Space When Generating Faces
Source
How to Explore the GAN Latent Space When Generating Faces (machinelearningmastery.com)(with code)
Key Takeaways
The latent space has structure that can be explored, such as by interpolating between points and performing vector arithmetic between points in latent space which have meaningful and targeted effects on the generated images.
Latent space visualization
Source
Latent space visualization — Deep Learning bits #2 | HackerNoon
Key Takeaways
In CAE(Convolutional Autoencoder architecture), the latent space contains a compressed representation of the image.
t-SNE visualization of MNIST(what level of detail the model is capable of extracting)
plotting the t-SNE embedding of MNIST from image space
2D projection of the latent space(Denser clusters)
probe the structure of the latent space: Linear interpolation
image space interpolation
latent space interpolation : We take the same start and end images and feed them to the encoder to obtain their latent space representation. We then interpolate between the two latent vectors, and feed these to the decoder.,( slowly, smoothly transform)
probe the structure of the latent space: vector arithmetics
