Timezone: »
Deep generative models are becoming widely used across science and industry for a variety of purposes. A common challenge is achieving a precise implicit or explicit representation of the data probability density. Recent proposals have suggested using classifier weights to refine the learned density of deep generative models. We extend this idea to all types of generative models and show how latent space refinement via iterated generative modeling can circumvent topological obstructions and improve precision. This methodology also applies to cases were the target model is non-differentiable and has many internal latent dimensions which must be marginalized over before refinement. We demonstrate our Latent Space Refinement (LaSeR) protocol on a variety of examples, focusing on the combinations of Normalizing Flows and Generative Adversarial Networks.
Author Information
Ramon Winterhalder (UC Louvain)
Marco Bellagente (Heidelberg University)
Benjamin Nachman (Lawrence Berkeley National Laboratory)
Related Events (a corresponding poster, oral, or spotlight)
-
2021 : Latent Space Refinement for Deep Generative Models »
Dates n/a. Room
More from the Same Authors
-
2021 : Classifying Anomalies THrough Outer Density Estimation (CATHODE) »
Joshua Isaacson · Gregor Kasieczka · Benjamin Nachman · David Shih · Manuel Sommerhalder -
2021 : Uncertainty Aware Learning for High Energy Physics With A Cautionary Tale »
Aishik Ghosh · Benjamin Nachman -
2021 : Symmetry Discovery with Deep Learning »
Krish Desai · Benjamin Nachman · Jesse Thaler -
2022 : Learning Uncertainties the Frequentist Way: Calibration and Correlation in High Energy Physics »
Rikab Gambhir · Jesse Thaler · Benjamin Nachman -
2022 : Efficiently Moving Instead of Reweighting Collider Events with Machine Learning »
Radha Mastandrea · Benjamin Nachman -
2022 : Deconvolving Detector Effects for Distribution Moments »
Krish Desai · Benjamin Nachman · Jesse Thaler -
2022 : Particle-level Compression for New Physics Searches »
Yifeng Huang · Jack Collins · Benjamin Nachman · Simon Knapen · Daniel Whiteson -
2022 : One-Class Dense Networks for Anomaly Detection »
Norman Karr · Benjamin Nachman · David Shih -
2022 : Anomaly Detection with Multiple Reference Datasets in High Energy Physics »
Mayee Chen · Benjamin Nachman · Frederic Sala -
2023 Workshop: NeurIPS 2023 Workshop: Machine Learning and the Physical Sciences »
Brian Nord · Atilim Gunes Baydin · Adji Bousso Dieng · Emine Kucukbenli · Siddharth Mishra-Sharma · Benjamin Nachman · Kyle Cranmer · Gilles Louppe · Savannah Thais -
2022 Workshop: Machine Learning and the Physical Sciences »
Atilim Gunes Baydin · Adji Bousso Dieng · Emine Kucukbenli · Gilles Louppe · Siddharth Mishra-Sharma · Benjamin Nachman · Brian Nord · Savannah Thais · Anima Anandkumar · Kyle Cranmer · Lenka Zdeborová · Rianne van den Berg -
2021 Workshop: Machine Learning and the Physical Sciences »
Anima Anandkumar · Kyle Cranmer · Mr. Prabhat · Lenka Zdeborová · Atilim Gunes Baydin · Juan Carrasquilla · Emine Kucukbenli · Gilles Louppe · Benjamin Nachman · Brian Nord · Savannah Thais -
2017 : Poster session 2 and coffee break »
Sean McGregor · Tobias Hagge · Markus Stoye · Trang Thi Minh Pham · Seungkyun Hong · Amir Farbin · Sungyong Seo · Susana Zoghbi · Daniel George · Stanislav Fort · Steven Farrell · Arthur Pajot · Kyle Pearson · Adam McCarthy · Cecile Germain · Dustin Anderson · Mario Lezcano Casado · Mayur Mudigonda · Benjamin Nachman · Luke de Oliveira · Li Jing · Lingge Li · Soo Kyung Kim · Timothy Gebhard · Tom Zahavy -
2017 : Poster session 1 and coffee break »
Tobias Hagge · Sean McGregor · Markus Stoye · Trang Thi Minh Pham · Seungkyun Hong · Amir Farbin · Sungyong Seo · Susana Zoghbi · Daniel George · Stanislav Fort · Steven Farrell · Arthur Pajot · Kyle Pearson · Adam McCarthy · Cecile Germain · Dustin Anderson · Mario Lezcano Casado · Mayur Mudigonda · Benjamin Nachman · Luke de Oliveira · Li Jing · Lingge Li · Soo Kyung Kim · Timothy Gebhard · Tom Zahavy