Reward driven discovery of the optimal microstructure representations with invariant variational autoencoders Article Swipe

PDF

Boris N. Slautin , Kamyar Barakati , Hiroshi Funakubo , Maxim Ziatdinov , Vladimir V. Shvartsman , Doru C. Lupascu , Sergei V. Kalinin ·

YOU? · · 2025 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2510.00243

Microscopy techniques generate vast amounts of complex image data that in principle can be used to discover simpler, interpretable, and parsimonious forms to reveal the underlying physical structures, such as elementary building blocks in molecular systems or order parameters and phases in crystalline materials. Variational Autoencoders (VAEs) provide a powerful means of constructing such low-dimensional representations, but their performance heavily depends on multiple non-myopic design choices, which are often optimized through trial-and-error and empirical analysis. To enable automated and unbiased optimization of VAE workflows, we investigated reward-based strategies for evaluating latent space representations. Using Piezoresponse Force Microscopy data as a model system, we examined multiple policies and reward functions that can serve as a foundation for automated optimization. Our analysis shows that approximating the latent space with Gaussian Mixture Models (GMM) and Bayesian Gaussian Mixture Models (BGMM) provides a strong basis for constructing reward functions capable of estimating model efficiency and guiding the search for optimal parsimonious representations.

Related Topics

A Discovery Of Witches (Tv Series)

Driven (2001 Film)

Discovery World (International Tv Channel)

Land Rover Discovery

Discovery (Daft Punk Album)

A Discovery Of Witches

Discovery Of The Tomb Of Tutankhamun

Space Shuttle Discovery

Discovery (Electric Light Orchestra Album)

Far Beyond Driven

Discovery, Inc.

Reward System

Concepts

No concepts available.

Metadata

Type: preprint
Language: en
Landing Page: http://arxiv.org/abs/2510.00243
PDF: https://arxiv.org/pdf/2510.00243
OA Status: green
OpenAlex ID: https://openalex.org/W4414806834

All OpenAlex metadata

Raw OpenAlex JSON

OpenAlex ID: https://openalex.org/W4414806834

Canonical identifier for this work in OpenAlex
DOI: https://doi.org/10.48550/arxiv.2510.00243

Digital Object Identifier
Title: Reward driven discovery of the optimal microstructure representations with invariant variational autoencoders

Work title
Type: preprint

OpenAlex work type
Language: en

Primary language
Publication year: 2025

Year of publication
Publication date: 2025-09-30

Full publication date if available
Authors: Boris N. Slautin, Kamyar Barakati, Hiroshi Funakubo, Maxim Ziatdinov, Vladimir V. Shvartsman, Doru C. Lupascu, Sergei V. Kalinin

List of authors in order
Landing page: https://arxiv.org/abs/2510.00243

Publisher landing page
PDF URL: https://arxiv.org/pdf/2510.00243

Direct link to full text PDF
Open access: Yes

Whether a free full text is available
OA status: green

Open access status per OpenAlex
OA URL: https://arxiv.org/pdf/2510.00243

Direct OA link when available
Cited by: 0

Total citation count in OpenAlex

Full payload

id	https://openalex.org/W4414806834
doi	https://doi.org/10.48550/arxiv.2510.00243
ids.doi	https://doi.org/10.48550/arxiv.2510.00243
ids.openalex	https://openalex.org/W4414806834
fwci
type	preprint
title	Reward driven discovery of the optimal microstructure representations with invariant variational autoencoders
biblio.issue
biblio.volume
biblio.last_page
biblio.first_page
topics[0].id	https://openalex.org/T11201
topics[0].field.id	https://openalex.org/fields/22
topics[0].field.display_name	Engineering
topics[0].score	0.9266999959945679
topics[0].domain.id	https://openalex.org/domains/3
topics[0].domain.display_name	Physical Sciences
topics[0].subfield.id	https://openalex.org/subfields/2211
topics[0].subfield.display_name	Mechanics of Materials
topics[0].display_name	Metallurgy and Material Forming
is_xpac	False
apc_list
apc_paid
language	en
locations[0].id	pmh:oai:arXiv.org:2510.00243
locations[0].is_oa	True
locations[0].source.id	https://openalex.org/S4306400194
locations[0].source.issn
locations[0].source.type	repository
locations[0].source.is_oa	True
locations[0].source.issn_l
locations[0].source.is_core	False
locations[0].source.is_in_doaj	False
locations[0].source.display_name	arXiv (Cornell University)
locations[0].source.host_organization	https://openalex.org/I205783295
locations[0].source.host_organization_name	Cornell University
locations[0].source.host_organization_lineage	https://openalex.org/I205783295
locations[0].license
locations[0].pdf_url	https://arxiv.org/pdf/2510.00243
locations[0].version	submittedVersion
locations[0].raw_type	text
locations[0].license_id
locations[0].is_accepted	False
locations[0].is_published	False
locations[0].raw_source_name
locations[0].landing_page_url	http://arxiv.org/abs/2510.00243
locations[1].id	doi:10.48550/arxiv.2510.00243
locations[1].is_oa	True
locations[1].source.id	https://openalex.org/S4306400194
locations[1].source.issn
locations[1].source.type	repository
locations[1].source.is_oa	True
locations[1].source.issn_l
locations[1].source.is_core	False
locations[1].source.is_in_doaj	False
locations[1].source.display_name	arXiv (Cornell University)
locations[1].source.host_organization	https://openalex.org/I205783295
locations[1].source.host_organization_name	Cornell University
locations[1].source.host_organization_lineage	https://openalex.org/I205783295
locations[1].license	cc-by
locations[1].pdf_url
locations[1].version
locations[1].raw_type	article
locations[1].license_id	https://openalex.org/licenses/cc-by
locations[1].is_accepted	False
locations[1].is_published
locations[1].raw_source_name
locations[1].landing_page_url	https://doi.org/10.48550/arxiv.2510.00243
indexed_in	arxiv, datacite
authorships[0].author.id	https://openalex.org/A5061740776
authorships[0].author.orcid	https://orcid.org/0000-0002-1237-7070
authorships[0].author.display_name	Boris N. Slautin
authorships[0].author_position	first
authorships[0].raw_author_name	Slautin, Boris N.
authorships[0].is_corresponding	False
authorships[1].author.id	https://openalex.org/A5043040873
authorships[1].author.orcid	https://orcid.org/0000-0003-3487-6652
authorships[1].author.display_name	Kamyar Barakati
authorships[1].author_position	middle
authorships[1].raw_author_name	Barakati, Kamyar
authorships[1].is_corresponding	False
authorships[2].author.id	https://openalex.org/A5023888355
authorships[2].author.orcid	https://orcid.org/0000-0002-1106-200X
authorships[2].author.display_name	Hiroshi Funakubo
authorships[2].author_position	middle
authorships[2].raw_author_name	Funakubo, Hiroshi
authorships[2].is_corresponding	False
authorships[3].author.id	https://openalex.org/A5013879711
authorships[3].author.orcid	https://orcid.org/0000-0003-2570-4592
authorships[3].author.display_name	Maxim Ziatdinov
authorships[3].author_position	middle
authorships[3].raw_author_name	Ziatdinov, Maxim A.
authorships[3].is_corresponding	False
authorships[4].author.id	https://openalex.org/A5058631175
authorships[4].author.orcid	https://orcid.org/0000-0002-7155-2473
authorships[4].author.display_name	Vladimir V. Shvartsman
authorships[4].author_position	middle
authorships[4].raw_author_name	Shvartsman, Vladimir V.
authorships[4].is_corresponding	False
authorships[5].author.id	https://openalex.org/A5042352170
authorships[5].author.orcid	https://orcid.org/0000-0002-6895-1334
authorships[5].author.display_name	Doru C. Lupascu
authorships[5].author_position	middle
authorships[5].raw_author_name	Lupascu, Doru C.
authorships[5].is_corresponding	False
authorships[6].author.id	https://openalex.org/A5048552375
authorships[6].author.orcid	https://orcid.org/0000-0001-5354-6152
authorships[6].author.display_name	Sergei V. Kalinin
authorships[6].author_position	last
authorships[6].raw_author_name	Kalinin, Sergei V.
authorships[6].is_corresponding	False
has_content.pdf	False
has_content.grobid_xml	False
is_paratext	False
open_access.is_oa	True
open_access.oa_url	https://arxiv.org/pdf/2510.00243
open_access.oa_status	green
open_access.any_repository_has_fulltext	False
created_date	2025-10-10T00:00:00
display_name	Reward driven discovery of the optimal microstructure representations with invariant variational autoencoders
has_fulltext	False
is_retracted	False
updated_date	2025-11-06T06:51:31.235846
primary_topic.id	https://openalex.org/T11201
primary_topic.field.id	https://openalex.org/fields/22
primary_topic.field.display_name	Engineering
primary_topic.score	0.9266999959945679
primary_topic.domain.id	https://openalex.org/domains/3
primary_topic.domain.display_name	Physical Sciences
primary_topic.subfield.id	https://openalex.org/subfields/2211
primary_topic.subfield.display_name	Mechanics of Materials
primary_topic.display_name	Metallurgy and Material Forming
cited_by_count	0
locations_count	2
best_oa_location.id	pmh:oai:arXiv.org:2510.00243
best_oa_location.is_oa	True
best_oa_location.source.id	https://openalex.org/S4306400194
best_oa_location.source.issn
best_oa_location.source.type	repository
best_oa_location.source.is_oa	True
best_oa_location.source.issn_l
best_oa_location.source.is_core	False
best_oa_location.source.is_in_doaj	False
best_oa_location.source.display_name	arXiv (Cornell University)
best_oa_location.source.host_organization	https://openalex.org/I205783295
best_oa_location.source.host_organization_name	Cornell University
best_oa_location.source.host_organization_lineage	https://openalex.org/I205783295
best_oa_location.license
best_oa_location.pdf_url	https://arxiv.org/pdf/2510.00243
best_oa_location.version	submittedVersion
best_oa_location.raw_type	text
best_oa_location.license_id
best_oa_location.is_accepted	False
best_oa_location.is_published	False
best_oa_location.raw_source_name
best_oa_location.landing_page_url	http://arxiv.org/abs/2510.00243
primary_location.id	pmh:oai:arXiv.org:2510.00243
primary_location.is_oa	True
primary_location.source.id	https://openalex.org/S4306400194
primary_location.source.issn
primary_location.source.type	repository
primary_location.source.is_oa	True
primary_location.source.issn_l
primary_location.source.is_core	False
primary_location.source.is_in_doaj	False
primary_location.source.display_name	arXiv (Cornell University)
primary_location.source.host_organization	https://openalex.org/I205783295
primary_location.source.host_organization_name	Cornell University
primary_location.source.host_organization_lineage	https://openalex.org/I205783295
primary_location.license
primary_location.pdf_url	https://arxiv.org/pdf/2510.00243
primary_location.version	submittedVersion
primary_location.raw_type	text
primary_location.license_id
primary_location.is_accepted	False
primary_location.is_published	False
primary_location.raw_source_name
primary_location.landing_page_url	http://arxiv.org/abs/2510.00243
publication_date	2025-09-30
publication_year	2025
referenced_works_count	0
abstract_inverted_index.a	48, 99, 113, 138
abstract_inverted_index.To	75
abstract_inverted_index.as	29, 98, 112
abstract_inverted_index.be	13
abstract_inverted_index.in	10, 33, 41
abstract_inverted_index.of	5, 51, 81, 146
abstract_inverted_index.on	61
abstract_inverted_index.or	36
abstract_inverted_index.to	15, 22
abstract_inverted_index.we	84, 102
abstract_inverted_index.Our	118
abstract_inverted_index.VAE	82
abstract_inverted_index.and	19, 39, 72, 78, 106, 131, 150
abstract_inverted_index.are	67
abstract_inverted_index.but	56
abstract_inverted_index.can	12, 110
abstract_inverted_index.for	88, 115, 141, 154
abstract_inverted_index.the	24, 123, 152
abstract_inverted_index.data	8, 97
abstract_inverted_index.such	28, 53
abstract_inverted_index.that	9, 109, 121
abstract_inverted_index.used	14
abstract_inverted_index.vast	3
abstract_inverted_index.with	126
abstract_inverted_index.(GMM)	130
abstract_inverted_index.Force	95
abstract_inverted_index.Using	93
abstract_inverted_index.basis	140
abstract_inverted_index.forms	21
abstract_inverted_index.image	7
abstract_inverted_index.means	50
abstract_inverted_index.model	100, 148
abstract_inverted_index.often	68
abstract_inverted_index.order	37
abstract_inverted_index.serve	111
abstract_inverted_index.shows	120
abstract_inverted_index.space	91, 125
abstract_inverted_index.their	57
abstract_inverted_index.which	66
abstract_inverted_index.(BGMM)	136
abstract_inverted_index.(VAEs)	46
abstract_inverted_index.Models	129, 135
abstract_inverted_index.blocks	32
abstract_inverted_index.design	64
abstract_inverted_index.enable	76
abstract_inverted_index.latent	90, 124
abstract_inverted_index.phases	40
abstract_inverted_index.reveal	23
abstract_inverted_index.reward	107, 143
abstract_inverted_index.search	153
abstract_inverted_index.strong	139
abstract_inverted_index.Mixture	128, 134
abstract_inverted_index.amounts	4
abstract_inverted_index.capable	145
abstract_inverted_index.complex	6
abstract_inverted_index.depends	60
abstract_inverted_index.guiding	151
abstract_inverted_index.heavily	59
abstract_inverted_index.optimal	155
abstract_inverted_index.provide	47
abstract_inverted_index.system,	101
abstract_inverted_index.systems	35
abstract_inverted_index.through	70
abstract_inverted_index.Bayesian	132
abstract_inverted_index.Gaussian	127, 133
abstract_inverted_index.analysis	119
abstract_inverted_index.building	31
abstract_inverted_index.choices,	65
abstract_inverted_index.discover	16
abstract_inverted_index.examined	103
abstract_inverted_index.generate	2
abstract_inverted_index.multiple	62, 104
abstract_inverted_index.physical	26
abstract_inverted_index.policies	105
abstract_inverted_index.powerful	49
abstract_inverted_index.provides	137
abstract_inverted_index.simpler,	17
abstract_inverted_index.unbiased	79
abstract_inverted_index.analysis.	74
abstract_inverted_index.automated	77, 116
abstract_inverted_index.empirical	73
abstract_inverted_index.functions	108, 144
abstract_inverted_index.molecular	34
abstract_inverted_index.optimized	69
abstract_inverted_index.principle	11
abstract_inverted_index.Microscopy	0, 96
abstract_inverted_index.efficiency	149
abstract_inverted_index.elementary	30
abstract_inverted_index.estimating	147
abstract_inverted_index.evaluating	89
abstract_inverted_index.foundation	114
abstract_inverted_index.materials.	43
abstract_inverted_index.non-myopic	63
abstract_inverted_index.parameters	38
abstract_inverted_index.strategies	87
abstract_inverted_index.techniques	1
abstract_inverted_index.underlying	25
abstract_inverted_index.workflows,	83
abstract_inverted_index.Variational	44
abstract_inverted_index.crystalline	42
abstract_inverted_index.performance	58
abstract_inverted_index.structures,	27
abstract_inverted_index.Autoencoders	45
abstract_inverted_index.constructing	52, 142
abstract_inverted_index.investigated	85
abstract_inverted_index.optimization	80
abstract_inverted_index.parsimonious	20, 156
abstract_inverted_index.reward-based	86
abstract_inverted_index.Piezoresponse	94
abstract_inverted_index.approximating	122
abstract_inverted_index.optimization.	117
abstract_inverted_index.interpretable,	18
abstract_inverted_index.low-dimensional	54
abstract_inverted_index.trial-and-error	71
abstract_inverted_index.representations,	55
abstract_inverted_index.representations.	92, 157
cited_by_percentile_year
countries_distinct_count	0
institutions_distinct_count	7
citation_normalized_percentile