Learning energy-based model with variational auto-encoder as amortized sampler

J Xie, Z Zheng, P Li - Proceedings of the AAAI Conference on Artificial …, 2021 - ojs.aaai.org
Due to the intractable partition function, training energy-based models (EBMs) by maximum
likelihood requires Markov chain Monte Carlo (MCMC) sampling to approximate the …

SDDM: score-decomposed diffusion models on manifolds for unpaired image-to-image translation

S Sun, L Wei, J Xing, J Jia… - … Conference on Machine …, 2023 - proceedings.mlr.press
Recent score-based diffusion models (SBDMs) show promising results in unpaired image-to-
image translation (I2I). However, existing methods, either energy-based or statistically …

Energy-guided entropic neural optimal transport

P Mokrov, A Korotin, A Kolesov, N Gushchin… - arXiv preprint arXiv …, 2023 - arxiv.org
Energy-based models (EBMs) are known in the Machine Learning community for decades.
Since the seminal works devoted to EBMs dating back to the noughties, there have been a …

SAC: Energy-Based Reinforcement Learning with Stein Soft Actor Critic

S Messaoud, B Mokeddem, Z Xue, L Pang, B An… - arXiv preprint arXiv …, 2024 - arxiv.org
Learning expressive stochastic policies instead of deterministic ones has been proposed to
achieve better stability, sample complexity, and robustness. Notably, in Maximum Entropy …

Rethinking cross-domain semantic relation for few-shot image generation

Y Gou, M Li, Y Lv, Y Zhang, Y Xing, Y He - Applied Intelligence, 2023 - Springer
Abstract Training well-performing Generative Adversarial Networks (GANs) with limited data
has always been challenging. Existing methods either require sufficient data (over 100 …

Cooperative training of fast thinking initializer and slow thinking solver for conditional learning

J Xie, Z Zheng, X Fang, SC Zhu… - IEEE Transactions on …, 2021 - ieeexplore.ieee.org
This paper studies the problem of learning the conditional distribution of a high-dimensional
output given an input, where the output and input may belong to two different domains, eg …

CoopInit: Initializing generative adversarial networks via cooperative learning

Y Zhao, J Xie, P Li - Proceedings of the AAAI Conference on Artificial …, 2023 - ojs.aaai.org
Numerous research efforts have been made to stabilize the training of the Generative
Adversarial Networks (GANs), such as through regularization and architecture design …

Energy-based continuous inverse optimal control

Y Xu, J Xie, T Zhao, C Baker, Y Zhao… - IEEE transactions on …, 2022 - ieeexplore.ieee.org
The problem of continuous inverse optimal control (over finite time horizon) is to learn the
unknown cost function over the sequence of continuous control variables from expert …

ASCFL: Accurate and speedy semi-supervised clustering federated learning

J He, B Gong, J Yang, H Wang, P Xu… - Tsinghua Science and …, 2023 - ieeexplore.ieee.org
The influence of non-Independent Identically Distribution (non-IID) data on Federated
Learning (FL) has been a serious concern. Clustered Federated Learning (CFL) is an …

Learning proposals for practical energy-based regression

FK Gustafsson, M Danelljan… - … Conference on Artificial …, 2022 - proceedings.mlr.press
Energy-based models (EBMs) have experienced a resurgence within machine learning in
recent years, including as a promising alternative for probabilistic regression. However …