Haizhao Yang

Haizhao Yang

Auto-generated profile, based on publicly available information

Haizhao Yang

Department of Mathematics University of Maryland College Park MD USA

Research Focus & Metrics

Academic Career Length

10Years (2017-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 January 2021 in SIAM Journal on Mathematical Analysis
Verified
Journal Article
Closed access
JLJianfeng Lu
Jianfeng Lu
ZSZuowei Shen
Zuowei Shen
HYHaizhao Yang
Haizhao Yang
SZShijun Zhang
Shijun Zhang
This paper establishes the optimal approximation error characterization of deep rectified linear unit (ReLU) networks for smooth functions in terms of both width and depth simultaneously. To that e...

1 September 2021 in Neural Networks
Verified
Journal Article
Closed access
ZSZuowei Shen
Zuowei Shen
HYHaizhao Yang
Haizhao Yang
SZShijun Zhang
Shijun Zhang

1 June 2021 in Journal of Computational Mathematics
Verified
Journal Article
Closed access
HMHadrien Montanelli
Hadrien Montanelli
HYHaizhao Yang
Haizhao Yang
QDQiang Du
Qiang Du
We prove a theorem concerning the approximation of generalized bandlimited multivariate functions by deep ReLU networks for which the curse of the dimensionality is overcome. Our theorem is based on a result by Maurey and on the ability of deep ReLU networks to approximate Chebyshev polynomials and analytic functions efficiently. Machine learning, Deep ReLU networks, Curse of dimensionality, Approximation theory, Bandlimited functions, Chebyshev polynomials. 68T01, 33F05, 41A10 COPYRIGHT: © Global Science Press hadrien.montanelli@polytechnique.edu (Hadrien Montanelli) haizhao@purdue.edu (Haizhao Yang) qd2125@columbia.edu (Qiang Du)

26 March 2021 in Neural Computation
Verified
Journal Article
Closed access
ZSZuowei Shen
Zuowei Shen
HYHaizhao Yang
Haizhao Yang
SZShijun Zhang
Shijun Zhang
A new network with super-approximation power is introduced. This network is built with Floor (⌊x⌋) or ReLU (max{0,x}) activation function in each neuron; hence, we call such networks Floor-ReLU networks. For any hyperparameters N∈N+ and L∈N+, we show that Floor-ReLU networks with width max{d,5N+13} and depth 64dL+3 can uniformly approximate a Hölder function f on [0,1]d with an approximation error 3λdα/2N-αL, where α∈(0,1] and λ are the Hölder order and constant, respectively. More generally for an arbitrary continuous function f on [0,1]d with a modulus of continuity ωf(·), the constructive approximation rate is ωf(dN-L)+2ωf(d)N-L. As a consequence, this new class of networks overcomes the curse of dimensionality in approximation power when the variation of ωf(r) as r→0 is moderate (e.g., ωf(r)≲rα for Hölder continuous functions), since the major term to be considered in our approximation rate is essentially d times a function of N and L independent of d within the modulus of continuity.

1 October 2020 in Journal of Computational Physics
Verified
Journal Article
Closed access
JHJianguo Huang
Jianguo Huang
HWHaoqin Wang
Haoqin Wang
HYHaizhao Yang
Haizhao Yang

1 March 2021 in Journal of Computational Physics
Verified
Journal Article
Closed access
JHJohn Harlim
John Harlim
SJShixiao Jiang
Shixiao Jiang
SLSenwei Liang
Senwei Liang
HYHaizhao Yang
Haizhao Yang