no code implementations • 23 Apr 2024 • Yingqing Guo, Hui Yuan, Yukang Yang, Minshuo Chen, Mengdi Wang
To remedy this issue, we consider a modified form of gradient guidance based on a forward prediction loss, which leverages the pre-trained score function to preserve the latent structure in generated samples.
no code implementations • 11 Apr 2024 • Minshuo Chen, Song Mei, Jianqing Fan, Mengdi Wang
In this paper, we review emerging applications of diffusion models, understanding their sample generation under various controls.
no code implementations • 20 Mar 2024 • Zihao Li, Hui Yuan, Kaixuan Huang, Chengzhuo Ni, Yinyu Ye, Minshuo Chen, Mengdi Wang
In this paper, we focus on diffusion models, a powerful generative AI technology, and investigate their potential for black-box optimization over complex structured variables.
no code implementations • 18 Mar 2024 • Hengyu Fu, Zhuoran Yang, Mengdi Wang, Minshuo Chen
Conditional diffusion models serve as the foundation of modern image synthesis and find extensive application in fields like computational biology and reinforcement learning.
no code implementations • 3 Mar 2024 • Yuchen Wu, Minshuo Chen, Zihao Li, Mengdi Wang, Yuting Wei
Diffusion models benefit from instillation of task-specific information into the score function to steer the sample generation towards desired properties.
no code implementations • 16 Oct 2023 • Zihao Li, Xiang Ji, Minshuo Chen, Mengdi Wang
In fact, human preference data are now used with classic reinforcement learning algorithms such as actor-critic methods, which involve evaluating an intermediate policy over a reward learned from human preference data with distribution shift, known as off-policy evaluation (OPE).
no code implementations • 25 Sep 2023 • Zhenghao Xu, Xiang Ji, Minshuo Chen, Mengdi Wang, Tuo Zhao
As a result, by properly choosing the network size and hyperparameters, NPMD can find an $\epsilon$-optimal policy with $\widetilde{O}(\epsilon^{-\frac{d}{\alpha}-2})$ samples in expectation, where $\alpha\in(0, 1]$ indicates the smoothness of environment.
no code implementations • 24 Jul 2023 • Xiang Ji, Huazheng Wang, Minshuo Chen, Tuo Zhao, Mengdi Wang
A popular approach is to utilize human feedback to learn a reward function for training.
no code implementations • 6 Jul 2023 • Jiacheng Guo, Minshuo Chen, Huan Wang, Caiming Xiong, Mengdi Wang, Yu Bai
This paper studies the sample-efficiency of learning in Partially Observable Markov Decision Processes (POMDPs), a challenging problem in reinforcement learning that is known to be exponentially hard in the worst-case.
no code implementations • 4 Jul 2023 • Kaiqi Zhang, Zixuan Zhang, Minshuo Chen, Yuma Takeda, Mengdi Wang, Tuo Zhao, Yu-Xiang Wang
Convolutional residual neural networks (ConvResNets), though overparameterized, can achieve remarkable prediction performance in practice, which cannot be well explained by conventional wisdom.
no code implementations • 26 Jun 2023 • Zixuan Zhang, Minshuo Chen, Mengdi Wang, Wenjing Liao, Tuo Zhao
Existing theories on deep nonparametric regression have shown that when the input data lie on a low-dimensional manifold, deep neural networks can adapt to the intrinsic data structures.
no code implementations • 2 Jun 2023 • Minshuo Chen, Jie Meng, Yu Bai, Yinyu Ye, H. Vincent Poor, Mengdi Wang
We present algorithms and establish near-optimal regret upper and lower bounds, of the form $\tilde{\mathcal{O}}(\sqrt{{\rm poly}(H) SAK})$, for RL in the delayed and missing observation settings.
no code implementations • 25 May 2023 • Shenghao Wu, Wenbin Zhou, Minshuo Chen, Shixiang Zhu
Estimating the counterfactual outcome of treatment is essential for decision-making in public health and clinical science, among others.
2 code implementations • 18 Mar 2023 • Qingru Zhang, Minshuo Chen, Alexander Bukharin, Nikos Karampatziakis, Pengcheng He, Yu Cheng, Weizhu Chen, Tuo Zhao
Therefore, many fine-tuning methods are proposed to learn incremental updates of pre-trained weights in a parameter efficient way, e. g., low-rank increments.
no code implementations • 25 Feb 2023 • Biraj Dahal, Alex Havrilla, Minshuo Chen, Tuo Zhao, Wenjing Liao
Many existing experiments have demonstrated that generative networks can generate high-dimensional complex data from a low-dimensional easy-to-sample distribution.
no code implementations • 14 Feb 2023 • Minshuo Chen, Kaixuan Huang, Tuo Zhao, Mengdi Wang
Furthermore, the generated distribution based on the estimated score function captures the data geometric structures and converges to a close vicinity of the data distribution.
no code implementations • 1 Dec 2022 • Jiahui Cheng, Minshuo Chen, Hao liu, Tuo Zhao, Wenjing Liao
Label Shift has been widely believed to be harmful to the generalization performance of machine learning models.
no code implementations • 9 Jun 2022 • Hao liu, Minshuo Chen, Siawpeng Er, Wenjing Liao, Tong Zhang, Tuo Zhao
Overparameterized neural networks enjoy great representation power on complex data, and more importantly yield sufficiently smooth output, which is crucial to their generalization and robustness.
no code implementations • 6 Jun 2022 • Xiang Ji, Minshuo Chen, Mengdi Wang, Tuo Zhao
We consider the off-policy evaluation problem of reinforcement learning using deep convolutional neural networks.
no code implementations • 4 May 2022 • Jie Wang, Minshuo Chen, Tuo Zhao, Wenjing Liao, Yao Xie
Based on the approximation theory of neural networks, we show that the neural network IPM test has the type-II risk in the order of $n^{-(s+\beta)/d}$, which is in the same order of the type-II risk as the H\"older IPM test.
no code implementations • 6 Jan 2022 • Siawpeng Er, Edward Liu, Minshuo Chen, Yan Li, Yuqi Liu, Tuo Zhao, Hua Wang
This paper presents a deep learning assisted synthesis approach for direct end-to-end generation of RF/mm-wave passive matching network with 3D EM structures.
no code implementations • 1 Jan 2022 • Hao liu, Haizhao Yang, Minshuo Chen, Tuo Zhao, Wenjing Liao
Learning operators between infinitely dimensional spaces is an important learning task arising in wide applications in machine learning, imaging science, mathematical modeling and simulations, etc.
1 code implementation • NeurIPS 2021 • Minshuo Chen, Yan Li, Ethan Wang, Zhuoran Yang, Zhaoran Wang, Tuo Zhao
Theoretically, under a weak coverage assumption that the experience dataset contains enough information about the optimal policy, we prove that for an episodic mean-field MDP with a horizon $H$ and $N$ training trajectories, SAFARI attains a sub-optimality gap of $\mathcal{O}(H^2d_{\rm eff} /\sqrt{N})$, where $d_{\rm eff}$ is the effective dimension of the function class for parameterizing the value function, but independent on the number of agents.
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 • 7 Sep 2021 • Hao liu, Minshuo Chen, Tuo Zhao, Wenjing Liao
Most of existing statistical theories on deep neural networks have sample complexities cursed by the data dimension and therefore cannot well explain the empirical success of deep learning on high-dimensional data.
1 code implementation • ACL 2021 • Chen Liang, Simiao Zuo, Minshuo Chen, Haoming Jiang, Xiaodong Liu, Pengcheng He, Tuo Zhao, Weizhu Chen
The Lottery Ticket Hypothesis suggests that an over-parametrized network consists of ``lottery tickets'', and training a certain collection of them (i. e., a subnetwork) can match the performance of the full model.
no code implementations • NeurIPS 2020 • Yujia Xie, Hanjun Dai, Minshuo Chen, Bo Dai, Tuo Zhao, Hongyuan Zha, Wei Wei, Tomas Pfister
Finding the k largest or smallest elements from a collection of scores, i. e., top-k operation, is an important model component widely used in information retrieval, machine learning, and data mining.
no code implementations • 3 Nov 2020 • Minshuo Chen, Hao liu, Wenjing Liao, Tuo Zhao
Our theory shows that deep neural networks are adaptive to the low-dimensional geometric structures of the covariates, and partially explains the success of deep learning for causal inference.
no code implementations • NeurIPS Workshop LMCA 2020 • Yujia Xie, Hanjun Dai, Minshuo Chen, Bo Dai, Tuo Zhao, Hongyuan Zha, Wei Wei, Tomas Pfister
The top-$k$ operation, i. e., finding the $k$ largest or smallest elements from a collection of scores, is an important model component, which is widely used in information retrieval, machine learning, and data mining.
no code implementations • 12 Oct 2020 • Yu Bai, Minshuo Chen, Pan Zhou, Tuo Zhao, Jason D. Lee, Sham Kakade, Huan Wang, Caiming Xiong
A common practice in meta-learning is to perform a train-validation split (\emph{train-val method}) where the prior adapts to the task on one split of the data, and the resulting predictor is evaluated on another split.
no code implementations • 25 Aug 2020 • David Munzer, Siawpeng Er, Minshuo Chen, Yan Li, Naga S. Mannem, Tuo Zhao, Hua Wang
We propose using machine learning models for the direct synthesis of on-chip electromagnetic (EM) passive structures to enable rapid or even automated designs and optimizations of RF/mm-Wave circuits.
no code implementations • NeurIPS 2020 • Minshuo Chen, Yu Bai, Jason D. Lee, Tuo Zhao, Huan Wang, Caiming Xiong, Richard Socher
When the trainable network is the quadratic Taylor model of a wide two-layer network, we show that neural representation can achieve improved sample complexities compared with the raw input: For learning a low-rank degree-$p$ polynomial ($p \geq 4$) in $d$ dimension, neural representation requires only $\tilde{O}(d^{\lceil p/2 \rceil})$ samples, while the best-known sample complexity upper bound for the raw input is $\tilde{O}(d^{p-1})$.
no code implementations • 16 Feb 2020 • Yujia Xie, Hanjun Dai, Minshuo Chen, Bo Dai, Tuo Zhao, Hongyuan Zha, Wei Wei, Tomas Pfister
The top-k operation, i. e., finding the k largest or smallest elements from a collection of scores, is an important model component, which is widely used in information retrieval, machine learning, and data mining.
no code implementations • 10 Feb 2020 • Minshuo Chen, Wenjing Liao, Hongyuan Zha, Tuo Zhao
Generative Adversarial Networks (GANs) have achieved a great success in unsupervised learning.
no code implementations • ICLR 2020 • Minshuo Chen, Yizhou Wang, Tianyi Liu, Zhuoran Yang, Xingguo Li, Zhaoran Wang, Tuo Zhao
Generative Adversarial Imitation Learning (GAIL) is a powerful and practical approach for learning sequential decision-making policies.
no code implementations • NeurIPS 2019 • Minshuo Chen, Haoming Jiang, Wenjing Liao, Tuo Zhao
The network size scales exponentially in the approximation error, with an exponent depending on the intrinsic dimension of the data and the smoothness of the function.
no code implementations • ICLR 2019 • Minshuo Chen, Xingguo Li, Tuo Zhao
We remark: (1) Our generalization bound for vanilla RNNs is significantly tighter than the best of existing results; (2) We are not aware of any other generalization bounds for MGU, LSTM, and Conv RNNs in the exiting literature; (3) We demonstrate the advantages of these variants in generalization.
no code implementations • NeurIPS 2019 • Tianyi Liu, Minshuo Chen, Mo Zhou, Simon S. Du, Enlu Zhou, Tuo Zhao
We show, however, that gradient descent combined with proper normalization, avoids being trapped by the spurious local optimum, and converges to a global optimum in polynomial time, when the weight of the first layer is initialized at 0, and that of the second layer is initialized arbitrarily in a ball.
no code implementations • NeurIPS 2019 • Minshuo Chen, Haoming Jiang, Wenjing Liao, Tuo Zhao
It therefore demonstrates the adaptivity of deep ReLU networks to low-dimensional geometric structures of data, and partially explains the power of deep ReLU networks in tackling high-dimensional data with low-dimensional geometric structures.
no code implementations • ICLR Workshop DeepGenStruct 2019 • Yujia Xie, Minshuo Chen, Haoming Jiang, Tuo Zhao, Hongyuan Zha
Optimal Transport (OT) naturally arises in many machine learning applications, yet the heavy computational burden limits its wide-spread uses.
no code implementations • ICLR 2019 • Haoming Jiang, Zhehui Chen, Minshuo Chen, Feng Liu, Dingding Wang, Tuo Zhao
Generative Adversarial Networks (GANs), though powerful, is hard to train.
no code implementations • 28 Dec 2018 • Haoming Jiang, Zhehui Chen, Minshuo Chen, Feng Liu, Dingding Wang, Tuo Zhao
Specifically, we propose a new reparameterization approach for the weight matrices of the discriminator in GANs, which allows us to directly manipulate the spectra of the weight matrices through various regularizers and constraints, without intensively computing singular value decompositions.
no code implementations • NeurIPS 2018 • Minshuo Chen, Lin Yang, Mengdi Wang, Tuo Zhao
Specifically, our goal is to estimate the principle component of time series data with respect to the covariance matrix of the stationary distribution.