Statistical Learning
BiSLS/SPS: Auto-tune Step Sizes for Stable Bi-level Optimization Chen Fan
This unified envelope strategy allows for the extension of the algorithms and their convergence guarantees to BO settings.
Learning Curves for Noisy Heterogeneous Feature-Subsampled Ridge Ensembles Benjamin S. Ruben
The random forest method is a popular example [3, 7].
Stable Nonconvex-Nonconcave Training via Linear Interpolation
By replacing the inner optimizer in RAPP we rediscover the family of Lookahead algorithms for which we establish convergence in cohypomonotone problems even when the base optimizer is taken to be gradient descent ascent.
Stable Nonconvex-Nonconcave Training via Linear Interpolation
By replacing the inner optimizer in RAPP we rediscover the family of Lookahead algorithms for which we establish convergence in cohypomonotone problems even when the base optimizer is taken to be gradient descent ascent.
Provably Fast Finite Particle Variants of SVGD via Virtual Particle Stochastic Approximation
Although the population (i.e, infinite-particle) limit dynamics of SVGD is well characterized, its behavior in the finite-particle regime is far less understood.
Hierarchical Semi-Implicit Variational Inference with Application to Diffusion Model Acceleration Longlin Y u
Moreover, given pre-trained score networks, HSIVI can be used to accelerate the sampling process of diffusion models with the score matching objective.
DiffKendall: A Novel Approach for Few-Shot Learning with Differentiable Kendall's Rank Correlation
Standard few-shot learning methods employ geometric similarity metrics such as cosine similarity and negative Euclidean distance to gauge the semantic relatedness between two features.
DiffKendall: A Novel Approach for Few-Shot Learning with Differentiable Kendall's Rank Correlation
Standard few-shot learning methods employ geometric similarity metrics such as cosine similarity and negative Euclidean distance to gauge the semantic relatedness between two features.