Unifying O(3) equivariant neural networks design with tensor-network formalism

Autor: Zimu Li, Zihan Pengmei, Han Zheng, Erik Thiede, Junyu Liu, Risi Kondor

Publikováno v: Machine Learning: Science and Technology, Vol 5, Iss 2, p 025044 (2024)

Many learning tasks, including learning potential energy surfaces from ab initio calculations, involve global spatial symmetries and permutational symmetry between atoms or general particles. Equivariant graph neural networks are a standard approach

Externí odkaz: https://doaj.org/article/8a01d8ecfa0d451d91b51ebe7e83e865

Multiresolution equivariant graph variational autoencoder

Autor: Truong Son Hy, Risi Kondor

Publikováno v: Machine Learning: Science and Technology, Vol 4, Iss 1, p 015031 (2023)

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 hi

Externí odkaz: https://doaj.org/article/e1a033e3a46c4df6bb2702bdf34db600

Deep learning for automated classification and characterization of amorphous materials

Autor: Kirk Swanson, Risi Kondor, Kyle Swanson, Joshua Lequieu, Shubhendu Trivedi

Publikováno v: Soft Matter. 16:435-446

It is difficult to quantify structure-property relationships and to identify structural features of complex materials. The characterization of amorphous materials is especially challenging because their lack of long-range order makes it difficult to

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::5c4778593e8177c8df915e8245a30f82
https://doi.org/10.1039/c9sm01903k

Speeding up Learning Quantum States through Group Equivariant Convolutional Quantum Ans\'atze

Autor: Han Zheng, Zimu Li, Junyu Liu, Sergii Strelchuk, Risi Kondor

Publikováno v: INSPIRE-HEP

We develop a theoretical framework for $S_n$-equivariant quantum convolutional circuits, building on and significantly generalizing Jordan's Permutational Quantum Computing (PQC) formalism. We show that quantum circuits are a natural choice for Fouri

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::e1a810b7716e15880892b272840f2e4b
http://arxiv.org/abs/2112.07611

A community-powered search of machine learning strategy space to find NMR property prediction models

Autor: Zico Kolter, Mordechai Kornbluth, Sanghoon Kim, Lars Andersen Bratholm, Risi Kondor, Jonathan P. Mailoa, Youngsoo Lee, Lam Dang, Goran Rakocevic, Brandon Anderson, Shaojie Bai, Will Gerrard, Milos R. Popovic, David R. Glowacki, Sunghwan Choi, Craig P. Butts, Thanh Tu Nguyen, Devin Willmott, Erik H. Thiede, Luka Stojanovic, Youhan Lee, Nebojsa Tijanic, Addison Howard, Wonho Song, Walter Reade, Andres Torrubia, Pavel Hanchar

Publikováno v: PLoS ONE, Vol 16, Iss 7, p e0253612 (2021)
Bratholm, L A, Gerrard, W, Anderson, B, Bai, S, Choi, S, Dang, L, Hanchar, P, Howard, A, Kim, S, Kolter, Z, Kondor, R, Kornbluth, M, Lee, Y, Lee, Y, Mailoa, J P, Nguyen, T T, Popovic, M, Rakocevic, G, Reade, W, Song, W, Stojanovic, L, Thiede, E H, Tijanic, N, Torrubia, A, Willmott, D, Butts, C P & Glowacki, D R 2021, ' A community-powered search of machine learning strategy space to find NMR property prediction models ', PLoS ONE, vol. 16, no. 7, e0253612 . https://doi.org/10.1371/journal.pone.0253612
PLoS ONE

The rise of machine learning (ML) has created an explosion in the potential strategies for using data to make scientific predictions. For physical scientists wishing to apply ML strategies to a particular domain, it can be difficult to assess in adva

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::672ab3f6c2ed91044cbf5995b3323ede
http://arxiv.org/abs/2008.05994