no code implementations • 9 Apr 2024 • Yuchen Zhu, Tianrong Chen, Evangelos A. Theodorou, Xie Chen, Molei Tao
This article considers the generative modeling of the states of quantum systems, and an approach based on denoising diffusion model is proposed.
no code implementations • 18 Mar 2024 • Lingkai Kong, Molei Tao
Explicit, momentum-based dynamics for optimizing functions defined on Lie groups was recently constructed, based on techniques such as variational optimization and left trivialization.
no code implementations • 27 Feb 2024 • Ye He, Kevin Rojas, Molei Tao
It first describes a framework, Diffusion Monte Carlo (DMC), based on the simulation of a denoising diffusion process with its score function approximated by a generic Monte Carlo estimator.
1 code implementation • 30 Nov 2023 • Alex Havrilla, Kevin Rojas, Wenjing Liao, Molei Tao
Diffusion generative models have achieved remarkable success in generating images with a fixed resolution.
no code implementations • 26 Oct 2023 • Yuqing Wang, Zhenghao Xu, Tuo Zhao, Molei Tao
This regularity, together with gradient descent using a large learning rate that favors flatter regions, results in these nontrivial dynamical behaviors.
1 code implementation • NeurIPS 2023 • Guan-Horng Liu, Tianrong Chen, Evangelos A. Theodorou, Molei Tao
In this work, we propose Mirror Diffusion Models (MDM), a new class of diffusion models that generate data on convex constrained sets without losing any tractability.
no code implementations • 25 Sep 2023 • Zihao Hu, Guanghui Wang, Xi Wang, Andre Wibisono, Jacob Abernethy, Molei Tao
In the context of Euclidean space, it is established that the last-iterates of both the extragradient (EG) and past extragradient (PEG) methods converge to the solution of monotone variational inequality problems at a rate of $O\left(\frac{1}{\sqrt{T}}\right)$ (Cai et al., 2022).
no code implementations • 30 Sep 2022 • Oswin So, Gongjie Li, Evangelos A. Theodorou, Molei Tao
Incorporating the Hamiltonian structure of physical dynamics into deep learning models provides a powerful way to improve the interpretability and prediction accuracy.
1 code implementation • 11 Jun 2022 • Qinsheng Zhang, Molei Tao, Yongxin Chen
In the CLD, a diffusion model by augmenting the diffusion process with velocity, our algorithm achieves an FID score of 2. 26, on CIFAR10, with only 50 number of score function evaluations~(NFEs) and an FID score of 2. 86 with only 27 NFEs.
no code implementations • 8 Jun 2022 • Andre Wibisono, Molei Tao, Georgios Piliouras
In this paper we study two-player bilinear zero-sum games with constrained strategy spaces.
1 code implementation • 27 May 2022 • Lingkai Kong, Yuqing Wang, Molei Tao
The problem of optimization on Stiefel manifold, i. e., minimizing functions of (not necessarily square) matrices that satisfy orthogonality constraints, has been extensively studied.
no code implementations • ICLR 2022 • Yuqing Wang, Minshuo Chen, Tuo Zhao, Molei Tao
Moreover, we rigorously establish an implicit bias of GD induced by such a large learning rate, termed 'balancing', meaning that magnitudes of $X$ and $Y$ at the limit of GD iterations will be close even if their initialization is significantly unbalanced.
no code implementations • 24 Sep 2021 • Ruilin Li, Molei Tao, Santosh S. Vempala, Andre Wibisono
The Mirror Langevin Diffusion (MLD) is a sampling analogue of mirror flow in continuous time, and it has nice convergence properties under log-Sobolev or Poincare inequalities relative to the Hessian metric, as shown by Chewi et al. (2020).
no code implementations • ICLR 2022 • Ruilin Li, Hongyuan Zha, Molei Tao
This article considers the popular MCMC method of unadjusted Langevin Monte Carlo (LMC) and provides a non-asymptotic analysis of its sampling error in 2-Wasserstein distance.
no code implementations • NeurIPS 2021 • Ruilin Li, Hongyuan Zha, Molei Tao
This bound improves the best previously known $\widetilde{\mathcal{O}}\left(\frac{d}{\epsilon}\right)$ result and is optimal in both dimension $d$ and accuracy tolerance $\epsilon$ for log-smooth and log-strongly-convex target measures.
no code implementations • 9 Mar 2021 • Renyi Chen, Molei Tao
For this special case, both generic approaches based on learning the vector field of the latent ODE and specialized approaches based on learning the Hamiltonian that generates the vector field exist.
no code implementations • NeurIPS 2020 • Kaixuan Huang, Yuqing Wang, Molei Tao, Tuo Zhao
We then compare the kernel of deep ResNets with that of deep FFNets and discover that the class of functions induced by the kernel of FFNets is asymptotically not learnable, as the depth goes to infinity.
no code implementations • 16 Jun 2020 • Ruilin Li, Hongyuan Zha, Molei Tao
Nesterov's Accelerated Gradient (NAG) for optimization has better performance than its continuous time limit (noiseless kinetic Langevin) when a finite step-size is employed \citep{shi2021understanding}.
no code implementations • 20 Feb 2020 • Ruilin Li, Xin Wang, Hongyuan Zha, Molei Tao
In our practical implementation of EWSG, the non-uniform subsampling is performed efficiently via a Metropolis-Hastings chain on the data index, which is coupled to the MCMC algorithm.
no code implementations • NeurIPS 2020 • Lingkai Kong, Molei Tao
This article suggests that deterministic Gradient Descent, which does not use any stochastic gradient approximation, can still exhibit stochastic behaviors.
no code implementations • 14 Feb 2020 • Kaixuan Huang, Yuqing Wang, Molei Tao, Tuo Zhao
We then compare the kernel of deep ResNets with that of deep FFNets and discover that the class of functions induced by the kernel of FFNets is asymptotically not learnable, as the depth goes to infinity.
no code implementations • 27 Jan 2020 • Molei Tao, Tomoki Ohsawa
The article considers smooth optimization of functions on Lie groups.