Generative AI Models

Welcome to the Generative AI Models repository.

Implemented Models

1. Variational Autoencoder (VAE)

The VAE is a generative model that learns to encode data into a latent space and decode it back to the original space, facilitating data generation.

Reference Paper:

• Auto-Encoding Variational Bayes — Diederik P. Kingma and Max Welling (ICLR 2014)

---

2. Vector Quantized Variational Autoencoder (VQ-VAE)

VQ-VAE introduces discrete latent variables using vector quantization, which addresses "posterior collapse" in traditional VAEs.

Reference Paper:

• Neural Discrete Representation Learning — Aaron van den Oord, Oriol Vinyals, and Koray Kavukcuoglu (NeurIPS 2017)

---

3. Attention-based VAE

This variant integrates attention mechanisms into the VAE framework to enhance the model's ability to focus on relevant parts of the input during encoding and decoding.

---

Models Under Development

4. Diffusion Models

Diffusion models are generative models that learn to generate data by reversing a gradual noising process, resulting in high-quality sample generation.

Reference Paper:

• Denoising Diffusion Probabilistic Models — Jonathan Ho, Ajay Jain, and Pieter Abbeel (NeurIPS 2020)

---

5. Consistency Models

Consistency models enable fast one-step generation without adversarial training by mapping noise directly to data.

Reference Paper:

• Consistency Models — Yang Song et al. (2023)

---

6. Flow Matching Models

Flow Matching models offer a simulation-free approach for training continuous normalizing flows by directly estimating a vector field that generates the data distribution.

Reference Paper:

• Flow Matching for Generative Modeling — Yaron Lipman et al. (ICLR 2023)

---

Repository Structure

Generative-AI-Models/
│
├── VAE/               # Variational Autoencoder Implementation
├── VQ-VAE/            # Vector Quantized VAE Implementation
├── Attention_VAE/     # Attention-based VAE Implementation
├── Diffusion_Models/  # (Planned) Diffusion Models
├── Consistency_Models/ # (Planned) Consistency Models
└── Flow_Matching_Models/ # (Planned) Flow Matching Models

---

Getting Started

1. Clone the repository:

   git clone https://github.com/Vozikis/Generative-AI-Models.git
   cd Generative-AI-Models

2. Navigate to the desired model's directory:

   cd VAE  # For Variational Autoencoder

---

References

• Kingma, D.P., & Welling, M. (2014). Auto-Encoding Variational Bayes. arXiv:1312.6114
• Van den Oord, A., Vinyals, O., & Kavukcuoglu, K. (2017). Neural Discrete Representation Learning. arXiv:1711.00937
• Ho, J., Jain, A., & Abbeel, P. (2020). Denoising Diffusion Probabilistic Models. arXiv:2006.11239
• Song, Y., et al. (2023). Consistency Models. arXiv:2303.01469
• Lipman, Y., et al. (2023). Flow Matching for Generative Modeling. arXiv:2210.02747