Zobrazeno 1 - 10
of 27
pro vyhledávání: '"Jiaoyang Huang"'
Autor:
Kenji Kawaguchi1 (AUTHOR) kawaguch@mit.edu, Jiaoyang Huang2 (AUTHOR) jiaoyang@math.harvard.edu, Kaelbling, Leslie Pack1 (AUTHOR) lpk@csail.mit.edu
Publikováno v:
Neural Computation. Dec2019, Vol. 31 Issue 12, p2293-2323. 31p. 4 Diagrams.
We consider Bayesian inference for large scale inverse problems, where computational challenges arise from the need for repeated evaluations of an expensive forward model. This renders most Markov chain Monte Carlo approaches infeasible, since they t
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::327bcb5719cf55b4ccb7c2a9101860ed
Publikováno v:
2021 IEEE International Conference on Big Data (Big Data).
Publikováno v:
Neural Computation. 31:1462-1498
In this paper, we analyze the effects of depth and width on the quality of local minima, without strong overparameterization and simplification assumptions in the literature. Without any simplification assumption, for deep nonlinear neural networks w
Autor:
Jiaoyang Huang
Publikováno v:
International Mathematics Research Notices. 2021:5898-5942
We study the $\beta $ analogue of the nonintersecting Poisson random walks. We derive a stochastic differential equation of the Stieltjes transform of the empirical measure process, which can be viewed as a dynamical version of the Nekrasov’s equat
Autor:
Benjamin Landon, Jiaoyang Huang
Publikováno v:
Probability Theory and Related Fields. 175:209-253
We study Dyson Brownian motion with general potential V and for general $$\beta \ge 1$$ . For short times $$t = o (1)$$ and under suitable conditions on V we obtain a local law and corresponding rigidity estimates on the particle locations; that is,
Publikováno v:
Ann. Probab. 48, no. 2 (2020), 916-962
We consider the statistics of the extreme eigenvalues of sparse random matrices, a class of random matrices that includes the normalized adjacency matrices of the Erdős–Rényi graph $G(N,p)$. Tracy–Widom fluctuations of the extreme eigenvalues f
Autor:
Benjamin Landon, Jiaoyang Huang
Publikováno v:
Ann. Inst. H. Poincaré Probab. Statist. 56, no. 1 (2020), 120-154
Nous nous interessons aux statistiques, dans l’interieur du spectre, des valeurs propres de matrices laplaciennes de grands graphes d’Erdős–Renyi aleatoires dans le regime ou $p\geq N^{\delta}/N$ pour un $\delta>0$ fixe arbitraire. Nous montro
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::694bf3060b8e715b0825c7ef503df884
https://projecteuclid.org/euclid.aihp/1580720485
https://projecteuclid.org/euclid.aihp/1580720485
Autor:
Jiaoyang Huang1 JH4427@NYU.EDU, Huang, Daniel Z.2 DZHUANG@CALTECH.EDU, Qing Yang3 YANGQ@USTC.EDU.CN, Guang Cheng4 GUANGCHENG@UCLA.EDU
Publikováno v:
Journal of Machine Learning Research. 2022, Vol. 23, p1-47. 47p.
Every Local Minimum Value Is the Global Minimum Value of Induced Model in Nonconvex Machine Learning
Publikováno v:
Neural computation. 31(12)
For nonconvex optimization in machine learning, this article proves that every local minimum achieves the globally optimal value of the perturbable gradient basis model at any differentiable point. As a result, nonconvex machine learning is theoretic