Useful Links
Computer Science
Artificial Intelligence
AI and Creativity
1. Foundations of AI and Creativity
2. Core Technologies for Generative AI
3. Applications in Creative Domains
4. Human-AI Collaboration in the Creative Process
5. Evaluation of AI-Generated Content
6. Philosophical and Ethical Implications
Core Technologies for Generative AI
Foundational Concepts in Generative Models
Probabilistic Modeling
Probability Distributions
Gaussian Distributions
Categorical Distributions
Mixture Models
Sampling Methods
Monte Carlo Sampling
Markov Chain Monte Carlo
Importance Sampling
Maximum Likelihood Estimation
Parameter Learning
Model Fitting
Overfitting Prevention
Latent Space Representation
Dimensionality Reduction
Principal Component Analysis
t-SNE
UMAP
Interpolation in Latent Space
Linear Interpolation
Spherical Interpolation
Semantic Interpolation
Disentangled Representations
Factor Separation
Controllable Generation
Interpretability
Training Data and Distribution
Data Collection and Curation
Web Scraping
Dataset Assembly
Quality Control
Data Preprocessing
Normalization
Augmentation
Filtering
Data Bias and Limitations
Selection Bias
Representation Gaps
Mitigation Strategies
Generative Adversarial Networks
Core Architecture
The Generator
Network Structure
Deconvolutional Layers
Upsampling Techniques
Skip Connections
Output Generation
Noise Input Processing
Feature Map Generation
Final Output Layer
The Discriminator
Network Structure
Convolutional Layers
Feature Extraction
Classification Head
Real vs Fake Classification
Binary Classification
Probability Scoring
Decision Boundaries
The Adversarial Training Process
Game Theory Foundation
Minimax Game
Nash Equilibrium
Zero-Sum Competition
Loss Functions
Generator Loss
Discriminator Loss
Alternative Loss Formulations
Training Instability and Solutions
Mode Collapse
Vanishing Gradients
Training Techniques
Progressive Growing
Spectral Normalization
Gradient Penalty
Key GAN Variants
StyleGAN
Style-Based Generation
Style Vectors
AdaIN Layers
Progressive Architecture
Applications in Portraits and Art
High-Resolution Faces
Artistic Style Transfer
Identity Manipulation
CycleGAN
Image-to-Image Translation
Domain Transfer
Style Conversion
Content Preservation
Unpaired Data Training
Cycle Consistency Loss
Bidirectional Translation
Domain Adaptation
Conditional GANs
Conditioning on Labels or Text
Class-Conditional Generation
Text-to-Image Synthesis
Attribute Control
Applications in Controlled Generation
Targeted Output
User-Specified Constraints
Interactive Generation
Progressive GANs
Incremental Resolution Training
Stability Improvements
High-Quality Output
Transformer Models
The Transformer Architecture
Self-Attention Mechanism
Query-Key-Value Framework
Attention Scores and Weights
Dot-Product Attention
Scaled Attention
Softmax Normalization
Multi-Head Attention
Parallel Attention Heads
Concatenation and Projection
Representation Diversity
Positional Encoding
Sinusoidal Encoding
Learned Positional Embeddings
Relative Position Encoding
Encoder-Decoder Structure
Encoder Functionality
Layer Stacking
Residual Connections
Layer Normalization
Decoder Functionality
Masked Self-Attention
Cross-Attention
Autoregressive Generation
Feed-Forward Networks
Position-wise Processing
Non-linear Transformations
Dimensionality Changes
Large Language Models
Model Scaling
Parameter Count Growth
Computational Requirements
Emergent Abilities
Pre-training and Fine-tuning
Unsupervised Pre-training
Next Token Prediction
Masked Language Modeling
Large-Scale Datasets
Transfer Learning
Knowledge Transfer
Task Adaptation
Few-Shot Learning
Domain Adaptation
Specialized Fine-tuning
Domain-Specific Data
Performance Optimization
Prompt Engineering
Prompt Design Strategies
Zero-Shot Prompting
Few-Shot Prompting
Chain-of-Thought Prompting
Prompt Tuning and Optimization
Soft Prompts
Prompt Search
Template Learning
Notable LLM Architectures
GPT Series
BERT and Variants
T5 and Text-to-Text Models
Diffusion Models
Theoretical Foundation
Stochastic Differential Equations
Reverse-Time Processes
Score-Based Models
The Forward Process
Noise Schedules
Linear Schedules
Cosine Schedules
Learned Schedules
Progressive Corruption
Gaussian Noise Addition
Variance Scheduling
Information Destruction
The Reverse Process
Denoising Steps
Iterative Refinement
Step Size Control
Convergence Criteria
Sampling Techniques
DDPM Sampling
DDIM Sampling
Accelerated Sampling
Neural Network Parameterization
U-Net Architecture
Time Conditioning
Noise Prediction
Conditioning and Guidance
Text-to-Image Generation
Text Embedding Integration
CLIP Embeddings
Cross-Attention Mechanisms
Semantic Alignment
Classifier-Free Guidance
Guidance Scaling
Unconditional Training
Quality-Diversity Trade-off
Image-to-Image Translation
Conditional Inputs
Inpainting Applications
Style Transfer
Advanced Diffusion Techniques
Latent Diffusion Models
Cascaded Diffusion
Continuous-Time Models
Other Key Architectures
Variational Autoencoders
Encoder-Decoder Structure
Probabilistic Encoding
Latent Variable Sampling
Reconstruction Decoding
Latent Variable Modeling
Prior Distributions
Posterior Approximation
KL Divergence Regularization
Applications in Generation
Continuous Latent Space
Interpolation Capabilities
Controllable Generation
Recurrent Neural Networks
Sequence Modeling
Temporal Dependencies
Hidden State Evolution
Variable-Length Sequences
Long Short-Term Memory
Memory Cells and Gates
Forget Gate
Input Gate
Output Gate
Applications in Sequential Data
Text Generation
Music Composition
Time Series Modeling
Gated Recurrent Units
Simplified Architecture
Reset and Update Gates
Computational Efficiency
Autoregressive Models
Sequential Generation
Conditional Probability Modeling
Causal Masking
Flow-Based Models
Normalizing Flows
Invertible Transformations
Exact Likelihood Computation
Previous
1. Foundations of AI and Creativity
Go to top
Next
3. Applications in Creative Domains