target distribution
The desired probability distribution that a learning algorithm aims to approximate during training, often in the context of generative models, guiding the sampling and optimization process.
- Assessing the quality of denoising diffusion models in Wasserstein distance: noisy score and optimal bounds
- Conditional Forecasts and Proper Scoring Rules for Reliable and Accurate Performative Predictions
- Convex Potential Mirror Langevin Algorithm for Efficient Sampling of Energy-Based Models
- Diffusion Generative Modeling on Lie Group Representations
- Fairness-aware Anomaly Detection via Fair Projection
- Implicit Generative Property Enhancer
- Non-equilibrium Annealed Adjoint Sampler
- The Adaptive Complexity of Minimizing Relative Fisher Information