Generative Models: An Interdisciplinary Perspective

被引:3
|
作者
Sankaran, Kris [1 ,2 ]
Holmes, Susan P. [3 ]
机构
[1] Univ Wisconsin, Wisconsin Inst Discovery, Madison, WI 53706 USA
[2] Univ Wisconsin, Dept Stat, Madison, WI 53706 USA
[3] Stanford Univ, Dept Stat, Stanford, CA 94305 USA
基金
美国国家科学基金会;
关键词
generative models; simulation; decision-making; agent based models; experimental design; particle filter; model evaluation; goodness-of-fit; QUALITY; DESIGN; SUBSET;
D O I
10.1146/annurev-statistics-033121-110134
中图分类号
O1 [数学];
学科分类号
0701 ; 070101 ;
摘要
By linking conceptual theories with observed data, generative models can support reasoning in complex situations. They have come to play a central role both within and beyond statistics, providing the basis for power analysis in molecular biology, theory building in particle physics, and resource allocation in epidemiology, for example. We introduce the probabilistic and computational concepts underlying modern generative models and then analyze how they can be used to inform experimental design, iterative model refinement, goodness-of-fit evaluation, and agent based simulation. We emphasize a modular view of generative mechanisms and discuss how they can be flexibly recombined in new problem contexts. We provide practical illustrations throughout, and code for reproducing all examples is available at https://github.com/krisrs1128/generative_review. Finally, we observe how research in generative models is currently split across several islands of activity, and we highlight opportunities lying at disciplinary intersections.
引用
收藏
页码:325 / 352
页数:28
相关论文
共 50 条
  • [1] Ethical Challenges and Solutions of Generative AI: An Interdisciplinary Perspective
    Al-kfairy, Mousa
    Mustafa, Dheya
    Kshetri, Nir
    Insiew, Mazen
    Alfandi, Omar
    INFORMATICS-BASEL, 2024, 11 (03):
  • [2] Generative Models for Relief Perspective Architectures
    Baglioni, Leonardo
    Fallavollita, Federico
    NEXUS NETWORK JOURNAL, 2021, 23 (04) : 879 - 898
  • [3] Generative Models for Relief Perspective Architectures
    Leonardo Baglioni
    Federico Fallavollita
    Nexus Network Journal, 2021, 23 : 879 - 898
  • [4] Explicit Disentanglement of Appearance and Perspective in Generative Models
    Detlefsen, Nicki S.
    Hauberg, Soren
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 32 (NIPS 2019), 2019, 32
  • [5] Generative Models from the perspective of Continual Learning
    Lesort, Timothee
    Caselles-Dupre, Hugo
    Garcia-Ortiz, Michael
    Stoian, Andrei
    Filliat, David
    2019 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN), 2019,
  • [6] Generative Models for Experimentation and Knowledge of Perspective Principles
    Fasolo, Marco
    Valenti, Graziano Mario
    Camagni, Flavia
    ADVANCES IN ADDITIVE MANUFACTURING, MODELING SYSTEMS AND 3D PROTOTYPING, 2020, 975 : 264 - 275
  • [7] ChatGPT and Beyond: Exploring the Responsible Use of Generative AI in the Workplace An Interdisciplinary Perspective
    Soellner, Matthias
    Arnold, Thomas
    Benlian, Alexander
    Bretschneider, Ulrich
    Knight, Caroline
    Ohly, Sandra
    Rudkowski, Lena
    Schreiber, Gerhard
    Wendt, Domenik
    BUSINESS & INFORMATION SYSTEMS ENGINEERING, 2025, 67 (02) : 289 - 303
  • [8] The Information Autoencoding Family: A Lagrangian Perspective on Latent Variable Generative Models
    Zhao, Shengjia
    Song, Jiaming
    Ermon, Stefano
    UNCERTAINTY IN ARTIFICIAL INTELLIGENCE, 2018, : 1031 - 1041
  • [9] A Variational Perspective on Diffusion-Based Generative Models and Score Matching
    Huang, Chin-Wei
    Lim, Jae Hyun
    Courville, Aaron
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 34 (NEURIPS 2021), 2021, 34
  • [10] Lesson planning models for GFL and GSL from an interdisciplinary perspective (II)
    Herzig, Katharina
    Ketzer-Noeltge, Almut
    DEUTSCH ALS FREMDSPRACHE-ZEITSCHRIFT ZUR THEORIE UND PRAXIS DES FACHES DEUTSCH ALS FREMDSPRACHE, 2020, 57 (01): : 27 - 40