Difference between revisions of "Generative AI"
| Line 8: | Line 8: | ||
[http://www.google.com/search?q=Generative+Modeling ...Google search] | [http://www.google.com/search?q=Generative+Modeling ...Google search] | ||
| − | |||
* [[AI Solver]] | * [[AI Solver]] | ||
| Line 16: | Line 15: | ||
* [http://en.wikipedia.org/wiki/Generative_model Generative model | Wikipedia] | * [http://en.wikipedia.org/wiki/Generative_model Generative model | Wikipedia] | ||
* [[Data Augmentation#Synthetic Labeling|Synthetic Labeling]] | * [[Data Augmentation#Synthetic Labeling|Synthetic Labeling]] | ||
| + | * Demos, generating... | ||
| + | ** Music: [http://colab.research.google.com/notebooks/magenta/piano_transformer/piano_transformer.ipynb Generating Piano Music with Transformer | I. Simon, A. Huang, J. Engel, C. "Fjord" Hawthorne - Google on Colab] play with pretrained [[Transformer]] models for piano music generation, based on the Music Transformer model | ||
| + | ** Faces: [http://colab.research.google.com/github/tensorflow/hub/blob/master/examples/colab/tf_hub_generative_image_module.ipynb TF-Hub generative image model | The TensorFlow Hub Authors - Google] use of a TF-Hub module based on a generative adversarial network (GAN). The module maps from N-dimensional vectors, called latent space, to RGB images. | ||
| + | ** 3D Objects: [http://colab.research.google.com/github/tensorflow/lucid/blob/master/notebooks/differentiable-parameterizations/style_transfer_3d.ipynb 3D Style Transfer | Google] uses Lucid to implement style transfer from a textured 3D model and a style image onto a new texture for the 3D model by using a [http://distill.pub/2018/differentiable-parameterizations/#section-style-transfer-3d Differentiable Image Parameterization]. | ||
| + | |||
| + | Generative modeling asks how likely it is, given condition X, that you’ll observe outcome Y. The approach has proved incredibly potent and versatile. [https://www.quantamagazine.org/how-artificial-intelligence-is-changing-science-20190311/ How Artificial Intelligence Is Changing Science | Dan Falk - Quanta Magazine] | ||
* Three main types: | * Three main types: | ||
| Line 28: | Line 33: | ||
*** [[Semi-Supervised Learning with Generative Adversarial Network (SSL-GAN)]] | *** [[Semi-Supervised Learning with Generative Adversarial Network (SSL-GAN)]] | ||
*** [[Context-Conditional Generative Adversarial Network (CC-GAN)]] | *** [[Context-Conditional Generative Adversarial Network (CC-GAN)]] | ||
| − | |||
| − | |||
| − | |||
<youtube>FZBFV7xfGaY</youtube> | <youtube>FZBFV7xfGaY</youtube> | ||
Revision as of 07:17, 2 October 2019
YouTube search... ...Google search
- AI Solver
- Generative Query Network (GQN)
- Discriminative vs. Generative
- Natural Language Generation (NLG)
- Generative model | Wikipedia
- Synthetic Labeling
- Demos, generating...
- Music: Generating Piano Music with Transformer | I. Simon, A. Huang, J. Engel, C. "Fjord" Hawthorne - Google on Colab play with pretrained Transformer models for piano music generation, based on the Music Transformer model
- Faces: TF-Hub generative image model | The TensorFlow Hub Authors - Google use of a TF-Hub module based on a generative adversarial network (GAN). The module maps from N-dimensional vectors, called latent space, to RGB images.
- 3D Objects: 3D Style Transfer | Google uses Lucid to implement style transfer from a textured 3D model and a style image onto a new texture for the 3D model by using a Differentiable Image Parameterization.
Generative modeling asks how likely it is, given condition X, that you’ll observe outcome Y. The approach has proved incredibly potent and versatile. How Artificial Intelligence Is Changing Science | Dan Falk - Quanta Magazine
- Three main types:
- Autoencoder (AE) / Encoder-Decoder
- Sequence Models
- Adversarial Networks
Deep Generative Modeling
Agents
Generative Modeling Language
