Amortizing intractable inference in diffusion models for vision, language, and control
Mila(米拉研究所);Université de Montréal(蒙特利尔大学);KAIST(韩国科学技术院);Jagiellonian University(雅盖隆大学);Ciela Institute(谢拉研究所);Dreamfold(Dreamfold公司);Recursion(Recursion公司);CIFAR(加拿大高等研究院);University of Edinburgh(爱丁堡大学)
Siddarth Venkatraman, Moksh Jain, Luca Scimeca, Minsu Kim, Marcin Sendera, Mohsin Hasan, Luke Rowe, Sarthak Mittal, Pablo Lemos, Emmanuel Bengio, Alexandre Adam… 展开作者
Siddarth Venkatraman, Moksh Jain, Luca Scimeca, Minsu Kim, Marcin Sendera, Mohsin Hasan, Luke Rowe, Sarthak Mittal, Pablo Lemos, Emmanuel Bengio, Alexandre Adam, Jarrid Rector-Brooks, Yoshua Bengio, Glen Berseth, Esmeralda S. Whitammer
Diffusion models have emerged as effective distribution estimators in vision, language, and reinforcement learning, but their use as priors in downstream tasks poses an intractable posterior inference problem. This paper studies amortized sampling of the posterior over data, $\mathbf{x}\sim p^{\rm post}(\mathbf{x})\propto p(\mathbf{x})r(\mathbf{x})$, in a model that consists of a diffusion generative model prior $p(\mathbf{x})$ and a black-box constraint or likelihood function $r(\mathbf{x})$. We state and prove the asymptotic correctness of a data-free learning objective, relative trajectory balance, for training a diffusion model that samples from this posterior, a problem that existing methods solve only approximately or in restricted cases. Relative trajectory balance arises from the generative flow network perspective on diffusion models, which allows the use of deep reinforcement learning techniques to improve mode coverage. Experiments illustrate the broad potential of unbiased inference of arbitrary posteriors under diffusion priors: in vision (classifier guidance), language (infilling under a discrete diffusion LLM), and multimodal data (text-to-image generation). Beyond generative modeling, we apply relative trajectory balance to the problem of continuous control with a score-based behavior prior, achieving state-of-the-art results on benchmarks in offline reinforcement learning.
Noise Hypernetworks: Amortizing Test-Time Compute in Diffusion Models
Technical University of Munich(慕尼黑技术大学);Munich Center of Machine Learning(慕尼黑机器学习中心);Helmholtz Munich(海德堡-慕尼黑 Helmholtz 中心);University of Tübingen(图宾根大学);Inceptive(Inceptive 公司);Google(谷歌公司)
South China University of Technology(华南理工大学);Guangzhou Laboratory(广州实验室);Tencent AI Lab(腾讯人工智能实验室)
联合分布匹配(Joint Distribution Matching,JDM)问题是指在机器学习和计算机视觉等应用中,通过学习双向映射来匹配两个域的联合分布的问题。然而,由于两个关键挑战,这个问题非常困难:(i)通常难以利用来自联合分布的足够信息来进行匹配;(ii)这个问题难以公式化和优化。在本文中,依靠最优运输理论,我们提出解决JDM问题的最小化Wasserstein距离的联合分布在两个区域。然而,由此产生的优化问题仍然是棘手的。在此基础上,我们提出了一个重要的定理,将这个复杂的问题转化为一个简单的优化问题,并提出了一种新的方法(称为联合Wasserstein分布匹配(JWDM))来解决这个问题。定性和定量的比较表明,我们的方法优于几个国家的最先进的性能。
$π_\texttt{RL}$: Online RL Fine-tuning for Flow-based Vision-Language-Action Models
Tsinghua University(清华大学);Peking University(北京大学);Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所);Carnegie Mellon University(卡内基梅隆大学);Infinigence AI;Zhongguancun Academy(中关村学院)
Bridging Generative and Discriminative Noisy-Label Learning via Direction-Agnostic EM Formulation
Cornell Tech(康奈尔科技);Cornell University(康奈尔大学);Department of Radiology, Stanford University(斯坦福大学放射科);Oxford University(牛津大学);Centre for Vision, Speech and Signal Processing (CVSSP), University of Surrey(视觉、语音和信号处理中心(CVSSP)、塞维利亚大学)
DEFUSE: Generalizable Backdoor Defense for Self-Supervised Encoders with Generative Priors
Ant Group(蚂蚁集团);Purple Mountain Laboratories(紫金山实验室);City University of Hong Kong(香港城市大学);Beijing Institute of Technology(北京理工大学);Alibaba Group(阿里巴巴集团)
The Perceptual Bandwidth Bottleneck in Vision-Language Models: Active Visual Reasoning via Sequential Experimental Design
The Hong Kong University of Science(香港科学与技术大学);University College London, London, United Kingdom(伦敦大学学院, 英国伦敦);ShanghaiTech University, Shanghai, China(上海科技大学, 中国上海);AI Lab, The Yangtze River Delta, China(人工智能实验室, 长江三角洲, 中国)
Simulation-based inference via telescoping ratio estimation for trawl processes
Statistics Program, Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division, King Abdullah University of Science and Technology (KAUST)(国王 Abdullah 科学与技术大学);Department of Mathematics, Imperial College London(伦敦帝国理工学院数学系)
Shenzhen Institute of Advanced Technology, CAS(中国科学院深圳先进技术研究院);The Chinese University of Hong Kong(香港中文大学);University of Edinburgh(爱丁堡大学);JD Explore Academy, JD.com, Inc(京东集团京东探索研究院);Tsinghua University(清华大学);The University of Sydney(悉尼大学)
Pro-t-VAE: Closed-Form Student's t Mixtures for Robust Prototype Discovery
Institute of Intelligent Information Processing, Shanxi University(山西大学智能信息处理研究所);School of Medicine, Yale University(耶鲁大学医学院);Boston Children’s Hospital, Harvard Medical School(哈佛医学院波士顿儿童医院);Eli Lilly and Company(礼来公司)