Generative AI Fundamentals
From GANs to diffusion to flow matching — build and customize generative AI systems
6 phases. 15 lessons. 15 labs. From GANs and VAEs through diffusion models, Stable Diffusion, and FLUX. Master ControlNet, LoRA, and DreamBooth for customization. Build video, audio, and 3D generation pipelines. Explore frontier techniques — flow matching, rectified flows, and visual autoregressive modeling (VAR). Evaluate with FID, CLIP score, and human preference. You build real generative pipelines and graduate with a portfolio that proves you can ship generative AI in production.
- Lessons: —
- Labs: —
- Projects: —
- Level: Beginner
Curriculum
- Generative AI Foundations — Generative model taxonomy (5 buckets), autoencoders, VAEs, KL divergence, the generative modeling framework
- GANs — Adversarial Generation — GANs, DCGANs, conditional GANs, Pix2Pix, StyleGAN, mode collapse, training stability
- Diffusion Models — DDPM to Stable Diffusion — DDPM from scratch, noise schedules, sampling (DDIM, DPM++), latent diffusion, Stable Diffusion, FLUX
- Control, Customization & Editing — ControlNet, LoRA, DreamBooth, textual inversion, inpainting, outpainting, image editing pipelines
- Multi-Modal Generation — Video generation (Sora, Veo, Kling), audio generation (Suno, MusicGen, Stable Audio), 3D generation (Gaussian Splatting, Tripo, Rodin)
- Frontier Techniques & Evaluation — Flow matching, rectified flows, visual autoregressive modeling (VAR), next-scale prediction, FID, CLIP score, human preference evaluation
Skills You Will Learn
- GANs from Scratch
- VAEs from Scratch
- DDPM (Denoising Diffusion Probabilistic Models)
- Latent Diffusion & Stable Diffusion / FLUX
- Text-to-Image Generation
- ControlNet for Conditional Generation
- LoRA Fine-Tuning
- DreamBooth & Textual Inversion
- Video, Audio & 3D Generation
- Flow Matching & Rectified Flows
- Visual Autoregressive Modeling (VAR)
- Generative Model Evaluation (FID, CLIP Score)
Related Courses
Browse all courses · View pricing · DeVenture Academy