approximation
In AI, approximation refers to the process of finding solutions that are close to the optimal solution, especially in contexts where exact solutions are computationally infeasible. It commonly appears in optimization problems and algorithm design, where approximations can yield sufficiently good results within acceptable error bounds.
- A Black-Box Debiasing Framework for Conditional Sampling
- Differentiable Generalized Sliced Wasserstein Plans
- Fully Dynamic Algorithms for Chamfer Distance
- Learning-Augmented Algorithms for $k$-median via Online Learning
- Monotone and Separable Set Functions: Characterizations and Neural Models
- Non-rectangular Robust MDPs with Normed Uncertainty Sets
- Scaling can lead to compositional generalization
- Shapley-Based Data Valuation for Weighted $k$-Nearest Neighbors