Generative modeling has recently seen many exciting developments with the advent of deep generative architectures such as Variational Auto-Encoders (VAE) or Generative Adversarial Networks (GAN). The ability to draw synthetic i.i.d. observations with the same joint probability distribution as a given dataset has a wide range of applications including representation learning, compression or imputation. It appears that it also has many applications in privacy preserving data analysis, especially when used in conjunction with differential privacy techniques. This paper focuses on synthetic data generation models with privacy preserving applications in mind. It introduces a novel architecture, the Composable Generative Model (CGM) that is state-of-the-art in tabular data generation. Any conditional generative model can be used as a sub-component of the CGM, including CGMs themselves, allowing the generation of numerical, categorical data as well as images, text, or time series. The CGM has been evaluated on 13 datasets (6 standard datasets and 7 simulated) and compared to 14 recent generative models. It beats the state of the art in tabular data generation by a significant margin.
Modeling Non-Ergodic Path Effects Using Conditional Generative Model for Fourier Amplitude Spectra
Department of Civil Engineering, University of California, Berkeley（加州大学伯克利分校土木工程系）；International Computer Science Institute, Berkeley, CA 94704, USA（伯克利计算机科学研究所）；Scientific Data Division, Lawrence Berkeley National Lab, Berkeley, CA 94720, USA（劳伦斯伯克利国家实验室科学数据部）；Energy Geosciences Division, Lawrence Berkeley National Lab, Berkeley, CA 94720, USA（劳伦斯伯克利国家实验室能源地球科学部）；Earthquake Research Institute, University of Tokyo, Tokyo 113-0032, Japan（东京大学地震研究所）；Department of Earth, Atmospheric and Planetary Sciences, MIT, Cambridge, MA 02139, USA（麻省理工学院地球、大气与行星科学系）；Department of Statistics, University of California, Berkeley, CA 94720, USA（加州大学伯克利分校统计系）
Zero-Shot Cyclic Peptide Design via Composable Geometric Constraints
Department of Computer Science and Technology, Tsinghua University（清华大学计算机科学与技术系）；Institute for AI Industry Research (AIR), Tsinghua（清华大学人工智能产业研究院）；Department of Electronic Engineering, Tsinghua University（清华大学电子工程系）；Xingjian College, Tsinghua University（清华大学xingjian学院）；Gaoling School of Artificial Intelligence, Renmin University of China（中国人民大学 Gallagher 人工智能学院）；Beijing Key Laboratory of Research on Large Models and Intelligent Governance（北京大模型与智能治理关键实验室）；Stanford University（斯坦福大学）
National University of Singapore（新加坡国立大学）；Institute for Infocomm Research, A*STAR（新加坡科技研究局资讯通信研究院）
This paper presents a novel collaborative generative modeling (CGM) framework that incentivizes collaboration among self-interested parties to contribute data to a pool for training a generative model (e.g., GAN), from which synthetic data are drawn and distributed to the parties as rewards commensurate to their contributions. Distributing synthetic data as rewards (instead of trained models or money) offers task- and model-agnostic benefits for downstream learning tasks and is less likely to violate data privacy regulation. To realize the framework, we firstly propose a data valuation functio…
University of Tübingen（图宾根大学）；Max Planck Institute for Intelligent Systems（马克斯·普朗克智能系统研究所）
Generative modeling of 3D brain MRIs presents difficulties in achieving high visual fidelity while ensuring sufficient coverage of the data distribution. In this work, we propose to address this challenge with composable, multiscale morphological transformations in a variational autoencoder (VAE) framework. These transformations are applied to a chosen reference brain image to generate MRI volumes, equipping the model with strong anatomical inductive biases. We structure the VAE latent space in a way such that the model covers the data distribution sufficiently well. We show substantial perfor…
Estimating the Hallucination Rate of Generative AI
Department of Statistics, Columbia University（统计系，哥伦比亚大学）；Department of Computer Science, Columbia University（计算机科学系，哥伦比亚大学）；OATML, Department of Computer Science, University of Oxford（OATML，计算机科学系，牛津大学）
Learning Physics for Unveiling Hidden Earthquake Ground Motions via Conditional Generative Modeling
Lawrence Berkeley National Lab（劳伦斯伯克利国家实验室）；Earthquake Research Institute, University of Tokyo（东京大学地震研究所）；University of California, Berkeley（加州大学伯克利分校）；Ben Gurion University of the Negev（本-古里安大学内盖夫分校）；MIT（麻省理工学院）；International Computer Science Institute（国际计算机科学研究所）
F$^3$low: Frame-to-Frame Coarse-grained Molecular Dynamics with SE(3) Guided Flow Matching
Mila - Québec AI Institute（米拉-魁北克人工智能研究所）；Université de Montréal（蒙特利尔大学）；Microsoft Research AI4Science（微软研究院AI4科学部门）；Xi’an Jiaotong University（西安交通大学）；Shanghai Jiao Tong University School of Medicine（上海交通大学医学院）；HEC Montréal（蒙特利尔高等商学院）；Canadian Institute for Advanced Research (CIFAR)（加拿大高级研究院）
From Glucose Patterns to Health Outcomes: A Generalizable Foundation Model for Continuous Glucose Monitor Data Analysis
Weizmann Institute of Science（魏茨曼科学研究所）；NVIDIA（英伟达公司）；Tel Aviv University（特拉维夫大学）；Schneider Children’s Medical Center of Israel（以色列施耐德儿童医疗中心）；University of Copenhagen（哥本哈根大学）；Massachusetts General Hospital（麻省总医院）；Garvan Institute of Medical Research（加文医学研究所）；University of NSW（新南威尔士大学）；St Vincent’s Hospital（圣文森特医院）；University of Zurich（苏黎世大学）；University of Santiago de Compostela（圣地亚哥德孔波斯特拉大学）；Mohamed bin Zayed University of Artificial Intelligence（穆罕默德·本·扎耶德人工智能大学）
Tel Aviv University（特拉维夫大学）；Simon Fraser University（西蒙弗雷泽大学）
Data-driven generative modeling has made remarkable progress by leveraging the power of deep neural networks. A reoccurring challenge is how to enable a model to generate a rich variety of samples from the entire target distribution, rather than only from a distribution confined to the training data. In other words, we would like the generative model to go beyond the observed samples and learn to generate ``unseen'', yet still plausible, data. In our work, we present CompoNet, a generative neural network for 2D or 3D shapes that is based on a part-based prior, where the key idea is for the net…
Diffusion Policies for Generative Modeling of Spacecraft Trajectories
Jet Propulsion Laboratory, California Institute of Technology（加州理工学院喷气推进实验室）；NASA Jet Propulsion Laboratory, California Institute of Technology（美国国家航空航天局加州理工学院喷气推进实验室）
