Semi-supervised Learning of Galaxy Morphology using Equivariant Transformer Variational Autoencoders
University of Oxford（牛津大学）
Mizu Nishikawa-Toomey, Lewis Smith, Yarin Gal
备注：
Accepted at the workshop for Machine Learning and the Physical Sciences, 34th Conference on Neural Information Processing Systems (NeurIPS) December 11, 2020
The growth in the number of galaxy images is much faster than the speed at which these galaxies can be labelled by humans. However, by leveraging the information present in the ever growing set of unlabelled images, semi-supervised learning could be an effective way of reducing the required labelling and increasing classification accuracy. We develop a Variational Autoencoder (VAE) with Equivariant Transformer layers with a classifier network from the latent space. We show that this novel architecture leads to improvements in accuracy when used for the galaxy morphology classification task on the Galaxy Zoo data set. In addition we show that pre-training the classifier network as part of the VAE using the unlabelled data leads to higher accuracy with fewer labels compared to exiting approaches. This novel VAE has the potential to automate galaxy morphology classification with reduced human labelling efforts.
In this paper, we propose Multiresolution Equivariant Graph Variational Autoencoders (MGVAE), the first hierarchical generative model to learn and generate graphs in a multiresolution and equivariant manner. At each resolution level, MGVAE employs higher order message passing to encode the graph while learning to partition it into mutually exclusive clusters and coarsening into a lower resolution that eventually creates a hierarchy of latent distributions. MGVAE then constructs a hierarchical generative model to variationally decode into a hierarchy of coarsened graphs. Importantly, our propos…
Beijing Institute for General Artificial Intelligence（北京通用人工智能研究院）；Tsinghua University（清华大学）；Peking University（北京大学）
Deep learning has achieved tremendous success in designing novel chemical compounds with desirable pharmaceutical properties. In this work, we focus on a new type of drug design problem -- generating a small "linker" to physically attach two independent molecules with their distinct functions. The main computational challenges include: 1) the generation of linkers is conditional on the two given molecules, in contrast to generating full molecules from scratch in previous works; 2) linkers heavily depend on the anchor atoms of the two molecules to be connected, which are not known beforehand; 3…
Variational Autoencoders (VAEs) have been shown to be remarkably effective in recovering model latent spaces for several computer vision tasks. However, currently trained VAEs, for a number of reasons, seem to fall short in learning invariant and equivariant clusters in latent space. Our work focuses on providing solutions to this problem and presents an approach to disentangle equivariance feature maps in a Lie group manifold by enforcing deep, group-invariant learning. Simultaneously implementing a novel separation of semantic and equivariant variables of the latent space representation, we …
Vrije Universiteit Amsterdam（阿姆斯特丹自由大学）；Qualcomm AI Research（高通人工智能研究院）；Qualcomm Technologies Netherlands B.V.（高通技术荷兰有限公司）；Qualcomm Technologies, Inc.（高通技术公司）
In compressed sensing, the goal is to reconstruct the signal from an underdetermined system of linear measurements. Thus, prior knowledge about the signal of interest and its structure is required. Additionally, in many scenarios, the signal has an unknown orientation prior to measurements. To address such recovery problems, we propose using equivariant generative models as a prior, which encapsulate orientation information in their latent space. Thereby, we show that signals with unknown orientations can be recovered with iterative gradient descent on the latent space of these models and prov…
This work addresses one-shot set and graph generation, and, more specifically, the parametrization of probabilistic decoders that map a vector-shaped prior to a distribution over sets or graphs. Sets and graphs are most commonly generated by first sampling points i.i.d. from a normal distribution, and then processing these points along with the prior vector using Transformer layers or Graph Neural Networks. This architecture is designed to generate exchangeable distributions, i.e., all permutations of the generated outputs are equally likely. We however show that it only optimizes a proxy to t…
Morphology-Aware Peptide Discovery via Masked Conditional Generative Modeling
Technical University of Munich（慕尼黑技术大学）；TUM School of Engineering and Design（慕尼黑技术大学工程与设计学院）；Department of Engineering Physics and Computation（工程物理与计算系）；Munich Data Science Institute（慕尼黑数据科学研究所）；AMC（原子模型中心）
Manipulating 3D Molecules in a Fixed-Dimensional E(3)-Equivariant Latent Space
Institute for AI Industry Research (AIR), Tsinghua University（人工智能产业研究院（AIR），清华大学）；Department of Computer Science and Technology, Tsinghua University（计算机科学与技术系，清华大学）；Qiuzhen College, Tsinghua University（启真学院，清华大学）
Deep learning (DL) methods where interpretability is intrinsically considered as part of the model are required to better understand the relationship of clinical and imaging-based attributes with DL outcomes, thus facilitating their use in the reasoning behind medical decisions. Latent space representations built with variational autoencoders (VAE) do not ensure individual control of data attributes. Attribute-based methods enforcing attribute disentanglement have been proposed in the literature for classical computer vision tasks in benchmark data. In this paper, we propose a VAE approach, th…
Pair-Variational Autoencoders (PairVAE) for Linking and Cross-Reconstruction of Characterization Data from Complementary Structural Characterization Techniques
University of Delaware（特拉华大学）
本文提出 PairVAE，利用配对 SAXS 和 SEM 数据训练变分自编码器，实现嵌段共聚物形态表征数据之间的交叉生成与重建，为二维 SAXS 图案解释和结构-性能关系数据库构建提供工具。
Multimodal Analysis of White Blood Cell Differentiation in Acute Myeloid Leukemia Patients using a β-Variational Autoencoder
Institute of AI for Health, Helmholtz Zentrum München – German Research Center for Environmental Health（慕尼黑亥姆霍兹中心人工智能健康研究所（德国环境健康研究中心））；University of Trento（特伦托大学）；Technical University of Munich（慕尼黑工业大学）；Ludwig-Maximilians University Munich（慕尼黑大学）；Johns Hopkins University（约翰斯·霍普金斯大学）
We present a conditional generative model to learn variation in cell and nuclear morphology and the location of subcellular structures from microscopy images. Our model generalizes to a wide range of subcellular localization and allows for a probabilistic interpretation of cell and nuclear morphology and structure localization from fluorescence images. We demonstrate the effectiveness of our approach by producing photo-realistic cell images using our generative model. The conditional nature of the model provides the ability to predict the localization of unobserved structures given cell and nu…
Hong Kong University of Science and Technology（香港科技大学）
Layout design with complex constraints is a challenging problem to solve due to the non-uniqueness of the solution and the difficulties in incorporating the constraints into the conventional optimization-based methods. In this paper, we propose a design method based on the recently developed machine learning technique, Variational Autoencoder (VAE). We utilize the learning capability of the VAE to learn the constraints and the generative capability of the VAE to generate design candidates that automatically satisfy all the constraints. As such, no constraints need to be imposed during the desi…
University of California at Irvine（加州大学欧文分校）；Columbia University（哥伦比亚大学）
While cloud-resolving models can explicitly simulate the details of small-scale storm formation and morphology, these details are often ignored by climate models for lack of computational resources. Here, we explore the potential of generative modeling to cheaply recreate small-scale storms by designing and implementing a Variational Autoencoder (VAE) that performs structural replication, dimensionality reduction, and clustering of high-resolution vertical velocity fields. Trained on ~6*10^6 samples spanning the globe, the VAE successfully reconstructs the spatial structure of convection, perf…
GLASS: Global Latent Aggregation with Slot-based Set Decoding for Scalable All-Atom Crystal Generation
Institute for Artificial Intelligence（人工智能研究所）；University of Stuttgart（斯图加特大学）；University of Toronto（多伦多大学）；Vector Institute（向量研究所）；NEC Labs Europe（NEC欧洲实验室）
A shape-similarity latent space for fluid interfaces: invertible reduced-order modelling of droplet morphology
Centre Internacional de Mètodes Numèrics en Enginyeria (CIMNE)（国际数值工程方法中心）；Universitat Politècnica de Catalunya – BarcelonaTech (UPC)（加泰罗尼亚理工大学）；Escola Tècnica Superior d’Enginyers de Camins, Canals i Ports, Universitat Politècnica de Catalunya – BarcelonaTech (UPC)（加泰罗尼亚理工大学土木、航道与港口工程学院）；Escola Politècnica Superior d’Enginyeria de Vilanova i la Geltrú (EPSEVG), Fluid Mechanics Department, Universitat Politècnica de Catalunya – BarcelonaTech (UPC)（加泰罗尼亚理工大学维尔诺瓦伊拉热特鲁高级工程学校）
