Zobraziť v EDS

Coupling CP with Deep Learning for Molecular Design and SARS-CoV2 Variants Exploration

Uložené v:
Podrobná bibliografia
Názov: Coupling CP with Deep Learning for Molecular Design and SARS-CoV2 Variants Exploration
Autori: Schiex, Thomas
Prispievatelia: Schiex, Thomas
Informácie o vydavateľovi: Array, 2023.
Rok vydania: 2023
Predmety: random Markov fields, [INFO.INFO-AI] Computer Science [cs]/Artificial Intelligence [cs.AI], constraint programming, Computing methodologies → Learning graphical models, deep learning, constraint programming, cost function networks, random Markov fields, decision-focused learning, protein design .CP.2023, decision-focused learning, 2012 ACM Subject Classification Computing methodologies → Artificial intelligence, deep learning, 2012 ACM Subject Classification Computing methodologies → Artificial intelligence Computing methodologies → Machine learning Theory of computation → Constraint and logic programming Computing methodologies → Learning graphical models, deep learning, constraint programming, cost function networks, random Markov fields, decision-focused learning, protein design .CP.2023, [INFO] Computer Science [cs], Computing methodologies → Artificial intelligence, [INFO.INFO-BT] Computer Science [cs]/Biotechnology, Computing methodologies → Machine learning, Computing methodologies → Learning in probabilistic graphical models, graphical models, cost function networks, Theory of computation → Constraint and logic programming, protein design, [INFO.INFO-BI] Computer Science [cs]/Bioinformatics [q-bio.QM]
Popis: The use of discrete optimization, including Constraint Programming, for designing objects that we completely understand is quite usual. In this talk, I'll show how designing specific biomolecules (proteins) raises new challenges, requiring solving problems that combine precise design targets, approximate laws, and design rules that can be deep-learned from data.
Druh dokumentu: Conference object
Popis súboru: application/pdf
Jazyk: English
DOI: 10.4230/lipics.cp.2023.4
Prístupová URL adresa: https://hal.inrae.fr/hal-04556854v1
https://hal.inrae.fr/hal-04556854v1/document
Rights: CC BY
Prístupové číslo: edsair.od......9730..f86bb9f99d5b0bbb4b322c984b387b9a
Databáza: OpenAIRE
Popis
Abstrakt:The use of discrete optimization, including Constraint Programming, for designing objects that we completely understand is quite usual. In this talk, I'll show how designing specific biomolecules (proteins) raises new challenges, requiring solving problems that combine precise design targets, approximate laws, and design rules that can be deep-learned from data.
DOI:10.4230/lipics.cp.2023.4