5 papers
Two-Sample Testing via Path-based Inference
Eshant English, Wei-Cheng Lai, Yanfeng Yang +3
Modern deep generative models are primarily studied for their ability to generate realistic samples, yet the generative dynamics they learn can also serve as objects of statistical…
Zero-Flow Two-Sample Tests
Yakun Wang, Leyang Wang, Song Liu +1
Motivated by the success of modern flow-based generative models in modeling complex data, we study two-sample testing through the lens of flow-based methods. We propose the Zero-Fl…
Zero-Flow Encoders
Yakun Wang, Leyang Wang, Song Liu +1
Flow-based methods have achieved significant success in various generative modeling tasks, capturing nuanced details within complex data distributions. However, few existing works…
In-Context Learning Is Provably Bayesian Inference: A Generalization Theory for Meta-Learning
Tomoya Wakayama, Taiji Suzuki
This paper develops a finite-sample statistical theory for in-context learning (ICL), analyzed within a meta-learning framework that accommodates mixtures of diverse task types. We…
Dimensionality-induced information loss of outliers in deep neural networks
Kazuki Uematsu, Kosuke Haruki, Taiji Suzuki +3
Out-of-distribution (OOD) detection is a critical issue for the stable and reliable operation of systems using a deep neural network (DNN). Although many OOD detection methods have…