Výsledky vyhledávání - "Jiang, Tianze"

Report

Prediction from compression for models with infinite memory, with applications to hidden Markov and renewal processes

Autor: Han, Yanjun, Jiang, Tianze, Wu, Yihong

Consider the problem of predicting the next symbol given a sample path of length n, whose joint distribution belongs to a distribution class that may have long-term memory. The goal is to compete with the conditional predictor that knows the true mod

Externí odkaz: http://arxiv.org/abs/2404.15454

Zobrazit plný text záznamu

Report

Density estimation using the perceptron

Autor: Gerber, Patrik Róbert, Jiang, Tianze, Polyanskiy, Yury, Sun, Rui

We propose a new density estimation algorithm. Given $n$ i.i.d. samples from a distribution belonging to a class of densities on $\mathbb{R}^d$, our estimator outputs any density in the class whose ''perceptron discrepancy'' with the empirical distri

Externí odkaz: http://arxiv.org/abs/2312.17701

Zobrazit plný text záznamu

Report

Kernel-Based Tests for Likelihood-Free Hypothesis Testing

Autor: Gerber, Patrik Róbert, Jiang, Tianze, Polyanskiy, Yury, Sun, Rui

Given $n$ observations from two balanced classes, consider the task of labeling an additional $m$ inputs that are known to all belong to \emph{one} of the two classes. Special cases of this problem are well-known: with complete knowledge of class dis

Externí odkaz: http://arxiv.org/abs/2308.09043

Zobrazit plný text záznamu

Report

Detection-Recovery and Detection-Refutation Gaps via Reductions from Planted Clique

Autor: Bresler, Guy, Jiang, Tianze

Planted Dense Subgraph (PDS) problem is a prototypical problem with a computational-statistical gap. It also exhibits an intriguing additional phenomenon: different tasks, such as detection or recovery, appear to have different computational limits.

Externí odkaz: http://arxiv.org/abs/2306.17719

Zobrazit plný text záznamu

Vyhledávací nástroje:

Upřesnit hledání