@article{19627, abstract = {Differentially private gradient descent (DP-GD) is a popular algorithm to train deep learning models with provable guarantees on the privacy of the training data. In the last decade, the problem of understanding its performance cost with respect to standard GD has received remarkable attention from the research community, which formally derived upper bounds on the excess population risk RP in different learning settings. However, existing bounds typically degrade with over-parameterization, i.e., as the number of parameters p gets larger than the number of training samples n -- a regime which is ubiquitous in current deep-learning practice. As a result, the lack of theoretical insights leaves practitioners without clear guidance, leading some to reduce the effective number of trainable parameters to improve performance, while others use larger models to achieve better results through scale. In this work, we show that in the popular random features model with quadratic loss, for any sufficiently large p , privacy can be obtained for free, i.e., |RP|=o(1) , not only when the privacy parameter ε has constant order, but also in the strongly private setting ε=o(1) . This challenges the common wisdom that over-parameterization inherently hinders performance in private learning.}, author = {Bombari, Simone and Mondelli, Marco}, issn = {1091-6490}, journal = {Proceedings of the National Academy of Sciences}, number = {15}, publisher = {National Academy of Sciences}, title = {{Privacy for free in the overparameterized regime}}, doi = {10.1073/pnas.2423072122}, volume = {122}, year = {2025}, } @article{19628, abstract = {We consider the critical temperature for superconductivity, defined via the linear BCS equation. We prove that at weak coupling the critical temperature for a sample confined to a quadrant in two dimensions is strictly larger than the one for a half-space, which in turn is strictly larger than the one for R^2. Furthermore, we prove that the relative difference of the critical temperatures vanishes in the weak coupling limit.}, author = {Roos, Barbara and Seiringer, Robert}, issn = {2050-5094}, journal = {Forum of Mathematics, Sigma}, publisher = {Cambridge University Press}, title = {{Enhanced superconductivity at a corner for the linear BCS equation}}, doi = {10.1017/fms.2024.145}, volume = {13}, year = {2025}, } @article{19629, abstract = {The SiOx anode exhibits a high specific capacity and commendable durability for lithium-ion batteries (LIBs). However, its practical application is hindered by significant volumetric fluctuations during lithiation/delithiation, alongside a metastable nature, which induces mechanical instability and irreversible lithium consumption, ultimately impairing long-term capacity retention in full-battery cell configurations. In this study, we present a phase-engineering approach designed to improve the structural stability of SiOx anodes for LIB applications. By incorporating lithium fluoride, amorphous SiOx undergoes partial transformation into a quartz-like phase, which enhances mechanical integrity and mitigates irreversible lithium loss. This modified anode demonstrates significantly improved stability and prolonged cycle lifespan. Through a combination of multiscale simulations and in situ characterizations, we elucidate the stabilization mechanisms conferred by the quartz phase, providing critical insights into the role of SiOx’s crystal structure in influencing degradation pathways. This work introduces an accessible and efficient method for controlling the crystallinity of SiOx, offering a practical solution to enhance the durability of high-energy-density LIBs.}, author = {Li, Jing and Zeng, Guifang and Horta, Sharona and Martínez-Alanis, Paulina R. and Jacas Biendicho, Jordi and Ibáñez, Maria and Xu, Bingang and Ci, Lijie and Cabot, Andreu and Sun, Qing}, issn = {1936-086X}, journal = {ACS Nano}, number = {16}, pages = {16096--16109}, publisher = {American Chemical Society}, title = {{Crystallographic engineering in micron-sized SiOx anode material toward stable high-energy-density Lithium-Ion batteries}}, doi = {10.1021/acsnano.5c03074}, volume = {19}, year = {2025}, } @phdthesis{19630, abstract = {This thesis consists of three chapters, each corresponding to one publication. While each of these projects tackles a topic in a different area of research, they all share a common thread in the type of topological structure they handle - a partition of space into volumes separated by interfaces that meet in non-manifold junctions. In Chapter 2, we study clusters of soap bubbles from a simulation perspective. In particular, we develop a surface-only algorithm that couples large scale motion and shape deformation of soap bubble clusters with the small scale evolution of the thin film's thickness, which is responsible for visual phenomena like surface vortices, Newton's interference patterns, capillary waves, and deformation-dependent rupturing of films in a foam. We model film thickness as a reduced degree of freedom in the Navier-Stokes equations and from them derive three sets of equations governing normal and tangential motion of the soap film surface, as well as the evolution of the thin film thickness. We discretize these equations on a non-manifold triangle mesh, extending and adapting operators to handle complex topology. We also present an incompressible fluid solver for 2.5D films and an advection algorithm for convecting fields across non-manifold surface junctions. Our simulations enhance bubble solvers with additional effects caused by convection, rippling, draining, and evaporation of the thin film. In Chapter 3, we introduce a multi-material non-manifold mesh-based surface tracking algorithm that converts mesh defects, such as overlaps, self-intersections, and inversions into topological changes. Our algorithm generalizes prior work on manifold surface tracking with topological changes: it preserves surface features like mesh-based methods, and it robustly handles topological changes like level set methods. Our method also offers improved efficiency and robustness over the state of the art. We demonstrate the effectiveness of the approach on a range of examples, including complex soap film simulations, such as those presented in Chapter 2, but with an order of magnitude more interacting bubbles than what we could achieve before, and Boolean unions of non-manifold meshes consisting of millions of triangles. Lastly, in Chapter 4, we utilize developments in the theory of random geometric complexes facilitated by observations from Discrete Morse theory. We survey the methods and results obtained with this new approach, and discuss some of its shortcomings. We use simulations to illustrate the results and to form conjectures, getting numerical estimates for combinatorial, topological, and geometric properties of weighted and unweighted Delaunay mosaics, their dual Voronoi tessellations, and the Alpha and Wrap complexes contained in the mosaics.}, author = {Synak, Peter}, issn = {2663-337X}, pages = {106}, publisher = {Institute of Science and Technology Austria}, title = {{Methods for fluid simulation, surface tracking, and statistics of non-manifold structures}}, doi = {10.15479/AT-ISTA-19630}, year = {2025}, } @article{19637, abstract = {PLATO (PLAnetary Transits and Oscillations of stars) is ESA’s M3 mission designed to detect and characterise extrasolar planets and perform asteroseismic monitoring of a large number of stars. PLATO will detect small planets (down to <2R Earth) around bright stars (<11 mag), including terrestrial planets in the habitable zone of solar-like stars. With the complement of radial velocity observations from the ground, planets will be characterised for their radius, mass, and age with high accuracy (5%, 10%, 10% for an Earth-Sun combination respectively). PLATO will provide us with a large-scale catalogue of well-characterised small planets up to intermediate orbital periods, relevant for a meaningful comparison to planet formation theories and to better understand planet evolution. It will make possible comparative exoplanetology to place our Solar System planets in a broader context. In parallel, PLATO will study (host) stars using asteroseismology, allowing us to determine the stellar properties with high accuracy, substantially enhancing our knowledge of stellar structure and evolution. The payload instrument consists of 26 cameras with 12cm aperture each. For at least four years, the mission will perform high-precision photometric measurements. Here we review the science objectives, present PLATO‘s target samples and fields, provide an overview of expected core science performance as well as a description of the instrument and the mission profile towards the end of the serial production of the flight cameras. PLATO is scheduled for a launch date end 2026. This overview therefore provides a summary of the mission to the community in preparation of the upcoming operational phases.}, author = {Rauer, Heike and Aerts, Conny and Cabrera, Juan and Deleuil, Magali and Erikson, Anders and Gizon, Laurent and Goupil, Mariejo and Heras, Ana and Walloschek, Thomas and Lorenzo-Alvarez, Jose and Marliani, Filippo and Martin-Garcia, César and Mas-Hesse, J. Miguel and O’Rourke, Laurence and Osborn, Hugh and Pagano, Isabella and Piotto, Giampaolo and Pollacco, Don and Ragazzoni, Roberto and Ramsay, Gavin and Udry, Stéphane and Appourchaux, Thierry and Benz, Willy and Brandeker, Alexis and Güdel, Manuel and Janot-Pacheco, Eduardo and Kabath, Petr and Kjeldsen, Hans and Min, Michiel and Santos, Nuno and Smith, Alan and Suarez, Juan Carlos and Werner, Stephanie C. and Aboudan, Alessio and Abreu, Manuel and Acuña, Lorena and Adams, Moritz and Adibekyan, Vardan and Affer, Laura and Agneray, François and Agnor, Craig and Aguirre Børsen-Koch, Victor and Ahmed, Saad and Aigrain, Suzanne and Al-Bahlawan, Ashraf and Alcacera Gil, Ma De Los Angeles and Alei, Eleonora and Alencar, Silvia and Alexander, Richard and Alfonso-Garzón, Julia and Alibert, Yann and Allende Prieto, Carlos and Almeida, Leonardo and Alonso Sobrino, Roi and Altavilla, Giuseppe and Althaus, Christian and Alvarez Trujillo, Luis Alonso and Amarsi, Anish and Ammler-Von Eiff, Matthias and Amôres, Eduardo and Andrade, Laerte and Antoniadis-Karnavas, Alexandros and António, Carlos and Aparicio Del Moral, Beatriz and Appolloni, Matteo and Arena, Claudio and Armstrong, David and Aroca Aliaga, Jose and Asplund, Martin and Audenaert, Jeroen and Auricchio, Natalia and Avelino, Pedro and Baeke, Ann and Baillié, Kevin and Balado, Ana and Ballber Balagueró, Pau and Balestra, Andrea and Ball, Warrick and Ballans, Herve and Ballot, Jerome and Barban, Caroline and Barbary, Gaële and Barbieri, Mauro and Barceló Forteza, Sebastià and Barker, Adrian and Barklem, Paul and Barnes, Sydney and Barrado Navascues, David and Barragan, Oscar and Baruteau, Clément and Basu, Sarbani and Baudin, Frederic and Baumeister, Philipp and Bayliss, Daniel and Bazot, Michael and Beck, Paul G. and Belkacem, Kevin and Bellinger, Earl and Benatti, Serena and Benomar, Othman and Bérard, Diane and Bergemann, Maria and Bergomi, Maria and Bernardo, Pierre and Biazzo, Katia and Bignamini, Andrea and Bigot, Lionel and Billot, Nicolas and Binet, Martin and Biondi, David and Biondi, Federico and Birch, Aaron C. and Bitsch, Bertram and Bluhm Ceballos, Paz Victoria and Bódi, Attila and Bognár, Zsófia and Boisse, Isabelle and Bolmont, Emeline and Bonanno, Alfio and Bonavita, Mariangela and Bonfanti, Andrea and Bonfils, Xavier and Bonito, Rosaria and Bonomo, Aldo Stefano and Börner, Anko and Boro Saikia, Sudeshna and Borreguero Martín, Elisa and Borsa, Francesco and Borsato, Luca and Bossini, Diego and Bouchy, Francois and Boué, Gwenaël and Boufleur, Rodrigo and Boumier, Patrick and Bourrier, Vincent and Bowman, Dominic M. and Bozzo, Enrico and Bradley, Louisa and Bray, John and Bressan, Alessandro and Breton, Sylvain and Brienza, Daniele and Brito, Ana and Brogi, Matteo and Brown, Beverly and Brown, David J.A. and Brun, Allan Sacha and Bruno, Giovanni and Bruns, Michael and Buchhave, Lars A. and Bugnet, Lisa Annabelle and Buldgen, Gaël and Burgess, Patrick and Busatta, Andrea and Busso, Giorgia and Buzasi, Derek and Caballero, José A. and Cabral, Alexandre and Cabrero Gomez, Juan Francisco and Calderone, Flavia and Cameron, Robert and Cameron, Andrew and Campante, Tiago and Campos Gestal, Néstor and Canto Martins, Bruno Leonardo and Cara, Christophe and Carone, Ludmila and Carrasco, Josep Manel and Casagrande, Luca and Casewell, Sarah L. and Cassisi, Santi and Castellani, Marco and Castro, Matthieu and Catala, Claude and Catalán Fernández, Irene and Catelan, Márcio and Cegla, Heather and Cerruti, Chiara and Cessa, Virginie and Chadid, Merieme and Chaplin, William and Charpinet, Stephane and Chiappini, Cristina and Chiarucci, Simone and Chiavassa, Andrea and Chinellato, Simonetta and Chirulli, Giovanni and Christensen-Dalsgaard, Jørgen and Church, Ross and Claret, Antonio and Clarke, Cathie and Claudi, Riccardo and Clermont, Lionel and Coelho, Hugo and Coelho, Joao and Cogato, Fabrizio and Colomé, Josep and Condamin, Mathieu and Conde García, Fernando and Conseil, Simon and Corbard, Thierry and Correia, Alexandre C.M. and Corsaro, Enrico and Cosentino, Rosario and Costes, Jean and Cottinelli, Andrea and Covone, Giovanni and Creevey, Orlagh L. and Crida, Aurelien and Csizmadia, Szilard and Cunha, Margarida and Curry, Patrick and Da Costa, Jefferson and Da Silva, Francys and Dalal, Shweta and Damasso, Mario and Damiani, Cilia and Damiani, Francesco and Das Chagas, Maria Liduina and Davies, Melvyn and Davies, Guy and Davies, Ben and Davison, Gary and De Almeida, Leandro and De Angeli, Francesca and De Barros, Susana Cristina Cabral and De Castroleão, Izan and De Freitas, Daniel Brito and De Freitas, Marcia Cristina and De Martino, Domitilla and De Medeiros, José Renan and De Paula, Luiz Alberto and De Pedraza Gómez, Álvaro and De Plaa, Jelle and De Ridder, Joris and Deal, Morgan and Decin, Leen and Deeg, Hans and Degl’Innocenti, Scilla and Deheuvels, Sebastien and Del Burgo, Carlos and Del Sordo, Fabio and Delgado-Mena, Elisa and Demangeon, Olivier and Denk, Tilmann and Derekas, Aliz and Desert, Jean Michel and Desidera, Silvano and Dexet, Marc and Di Criscienzo, Marcella and Di Giorgio, Anna Maria and Di Mauro, Maria Pia and Diaz Rial, Federico Jose and Díaz-García, José Javier and Dima, Marco and Dinuzzi, Giacomo and Dionatos, Odysseas and Distefano, Elisa and Do Nascimento, Jose Dias and Domingo, Albert and D’Orazi, Valentina and Dorn, Caroline and Doyle, Lauren and Duarte, Elena and Ducellier, Florent and Dumaye, Luc and Dumusque, Xavier and Dupret, Marc Antoine and Eggenberger, Patrick and Ehrenreich, David and Eigmüller, Philipp and Eising, Johannes and Emilio, Marcelo and Eriksson, Kjell and Ermocida, Marco and Escate Giribaldi, Riano Isidoro and Eschen, Yoshi and Espinosa Yáñez, Lucía and Estrela, Inês and Evans, Dafydd Wyn and Fabbian, Damian and Fabrizio, Michele and Faria, João Pedro and Farina, Maria and Farinato, Jacopo and Feliz, Dax and Feltzing, Sofia and Fenouillet, Thomas and Fernández, Miguel and Ferrari, Lorenza and Ferraz-Mello, Sylvio and Fialho, Fabio and Fienga, Agnes and Figueira, Pedro and Fiori, Laura and Flaccomio, Ettore and Focardi, Mauro and Foley, Steve and Fontignie, Jean and Ford, Dominic and Fornazier, Karin and Forveille, Thierry and Fossati, Luca and Franca, Rodrigo De Marca and Franco Da Silva, Lucas and Frasca, Antonio and Fridlund, Malcolm and Furlan, Marco and Gabler, Sarah Maria and Gaido, Marco and Gallagher, Andrew and Gallego Sempere, Paloma I. and Galli, Emanuele and García, Rafael A. and García Hernández, Antonio and Garcia Munoz, Antonio and García-Vázquez, Hugo and Garrido Haba, Rafael and Gaulme, Patrick and Gauthier, Nicolas and Gehan, Charlotte and Gent, Matthew and Georgieva, Iskra and Ghigo, Mauro and Giana, Edoardo and Gill, Samuel and Girardi, Leo and Giuliatti Winter, Silvia and Giusi, Giovanni and Gomes Da Silva, João and Gómez Zazo, Luis Jorge and Gomez-Lopez, Juan Manuel and González Hernández, Jonay Isai and Gonzalez Murillo, Kevin and Gonzalo Melchor, Alejandro and Gorius, Nicolas and Gouel, Pierre Vincent and Goulty, Duncan and Granata, Valentina and Grenfell, John Lee and Grießbach, Denis and Grolleau, Emmanuel and Grouffal, Salomé and Grziwa, Sascha and Guarcello, Mario Giuseppe and Gueguen, Loïc and Guenther, Eike Wolf and Guilhem, Terrasa and Guillerot, Lucas and Guillot, Tristan and Guiot, Pierre and Guterman, Pascal and Gutiérrez, Antonio and Gutiérrez-Canales, Fernando and Hagelberg, Janis and Haldemann, Jonas and Hall, Cassandra and Handberg, Rasmus and Harrison, Ian and Harrison, Diana L. and Hasiba, Johann and Haswell, Carole A. and Hatalova, Petra and Hatzes, Artie and Haywood, Raphaelle and Hébrard, Guillaume and Heckes, Frank and Heiter, Ulrike and Hekker, Saskia and Heller, René and Helling, Christiane and Helminiak, Krzysztof and Hemsley, Simon and Heng, Kevin and Herbst, Konstantin and Hermans, Aline and Hermes, J. J. and Hidalgo Torres, Nadia and Hinkel, Natalie and Hobbs, David and Hodgkin, Simon and Hofmann, Karl and Hojjatpanah, Saeed and Houdek, Günter and Huber, Daniel and Huesler, Joseph and Hui-Bon-Hoa, Alain and Huygen, Rik and Huynh, Duc Dat and Iro, Nicolas and Irwin, Jonathan and Irwin, Mike and Izidoro, André and Jacquinod, Sophie and Jannsen, Nicholas Emborg and Janson, Markus and Jeszenszky, Harald and Jiang, Chen and Jimenez Mancebo, Antonio José and Jofre, Paula and Johansen, Anders and Johnston, Cole and Jones, Geraint and Kallinger, Thomas and Kálmán, Szilárd and Kanitz, Thomas and Karjalainen, Marie and Karjalainen, Raine and Karoff, Christoffer and Kawaler, Steven and Kawata, Daisuke and Keereman, Arnoud and Keiderling, David and Kennedy, Tom and Kenworthy, Matthew and Kerschbaum, Franz and Kidger, Mark and Kiefer, Flavien and Kintziger, Christian and Kislyakova, Kristina and Kiss, László and Klagyivik, Peter and Klahr, Hubert and Klevas, Jonas and Kochukhov, Oleg and Köhler, Ulrich and Kolb, Ulrich and Koncz, Alexander and Korth, Judith and Kostogryz, Nadiia and Kovács, Gábor and Kovács, József and Kozhura, Oleg and Krivova, Natalie and Kuĉinskas, Arūnas and Kuhlemann, Ilyas and Kupka, Friedrich and Laauwen, Wouter and Labiano, Alvaro and Lagarde, Nadege and Laget, Philippe and Laky, Gunter and Lam, Kristine Wai Fun and Lambrechts, Michiel and Lammer, Helmut and Lanza, Antonino Francesco and Lanzafame, Alessandro and Lares Martiz, Mariel and Laskar, Jacques and Latter, Henrik and Lavanant, Tony and Lawrenson, Alastair and Lazzoni, Cecilia and Lebre, Agnes and Lebreton, Yveline and Lecavelier Des Etangs, Alain and Lee, Katherine and Leinhardt, Zoe and Leleu, Adrien and Lendl, Monika and Leto, Giuseppe and Levillain, Yves and Libert, Anne Sophie and Lichtenberg, Tim and Ligi, Roxanne and Lignieres, Francois and Lillo-Box, Jorge and Linsky, Jeffrey and Liu, John Scige and Loidolt, Dominik and Longval, Yuying and Lopes, Ilídio and Lorenzani, Andrea and Ludwig, Hans Guenter and Lund, Mikkel and Lundkvist, Mia Sloth and Luri, Xavier and Maceroni, Carla and Madden, Sean and Madhusudhan, Nikku and Maggio, Antonio and Magliano, Christian and Magrin, Demetrio and Mahy, Laurent and Maibaum, Olaf and Malac-Allain, Lee Roy and Malapert, Jean Christophe and Malavolta, Luca and Maldonado, Jesus and Mamonova, Elena and Manchon, Louis and Manjón, Andres and Mann, Andrew and Mantovan, Giacomo and Marafatto, Luca and Marconi, Marcella and Mardling, Rosemary and Marigo, Paola and Marinoni, Silvia and Marques, Rico and Marques, Joao Pedro and Marrese, Paola Maria and Marshall, Douglas and Martínez Perales, Silvia and Mary, David and Marzari, Francesco and Masana, Eduard and Mascher, Andrina and Mathis, Stéphane and Mathur, Savita and Martín Vodopivec, Iris and Mattiuci Figueiredo, Ana Carolina and Maxted, Pierre F.L. and Mazeh, Tsevi and Mazevet, Stephane and Mazzei, Francesco and Mccormac, James and Mcmillan, Paul and Menou, Lucas and Merle, Thibault and Meru, Farzana and Mesa, Dino and Messina, Sergio and Mészáros, Szabolcs and Meunier, Nadége and Meunier, Jean Charles and Micela, Giuseppina and Michaelis, Harald and Michel, Eric and Michielsen, Mathias and Michtchenko, Tatiana and Miglio, Andrea and Miguel, Yamila and Milligan, David and Mirouh, Giovanni and Mitchell, Morgan and Moedas, Nuno and Molendini, Francesca and Molnár, László and Mombarg, Joey and Montalban, Josefina and Montalto, Marco and Monteiro, Mário J.P.F.G. and Montoro Sánchez, Francisco and Morales, Juan Carlos and Morales-Calderon, Maria and Morbidelli, Alessandro and Mordasini, Christoph and Moreau, Chrystel and Morel, Thierry and Morello, Giuseppe and Morin, Julien and Mortier, Annelies and Mosser, Benoît and Mourard, Denis and Mousis, Olivier and Moutou, Claire and Mowlavi, Nami and Moya, Andrés and Muehlmann, Prisca and Muirhead, Philip and Munari, Matteo and Musella, Ilaria and Mustill, Alexander James and Nardetto, Nicolas and Nardiello, Domenico and Narita, Norio and Nascimbeni, Valerio and Nash, Anna and Neiner, Coralie and Nelson, Richard P. and Nettelmann, Nadine and Nicolini, Gianalfredo and Nielsen, Martin and Niemi, Sami Matias and Noack, Lena and Noels-Grotsch, Arlette and Noll, Anthony and Norazman, Azib and Norton, Andrew J. and Nsamba, Benard and Ofir, Aviv and Ogilvie, Gordon and Olander, Terese and Olivetto, Christian and Olofsson, Göran and Ong, Joel and Ortolani, Sergio and Oshagh, Mahmoudreza and Ottacher, Harald and Ottensamer, Roland and Ouazzani, Rhita Maria and Paardekooper, Sijme Jan and Pace, Emanuele and Pajas, Miriam and Palacios, Ana and Palandri, Gaelle and Palle, Enric and Paproth, Carsten and Parro, Vanderlei and Parviainen, Hannu and Pascual Granado, Javier and Passegger, Vera Maria and Pastor-Morales, Carmen and Pätzold, Martin and Pedersen, May Gade and Pena Hidalgo, David and Pepe, Francesco and Pereira, Filipe and Persson, Carina M. and Pertenais, Martin and Peter, Gisbert and Petit, Antoine C. and Petit, Pascal and Pezzuto, Stefania and Pichierri, Gabriele and Pietrinferni, Adriano and Pinheiro, Fernando and Pinsonneault, Marc and Plachy, Emese and Plasson, Philippe and Plez, Bertrand and Poppenhaeger, Katja and Poretti, Ennio and Portaluri, Elisa and Portell, Jordi and Porto De Mello, Gustavo Frederico and Poyatos, Julien and Pozuelos, Francisco J. and Prada Moroni, Pier Giorgio and Pricopi, Dumitru and Prisinzano, Loredana and Quade, Matthias and Quirrenbach, Andreas and Rabanal Reina, Julio Arturo and Rabello Soares, Maria Cristina and Raimondo, Gabriella and Rainer, Monica and Ramón Rodón, Jose and Ramón-Ballesta, Alejandro and Ramos Zapata, Gonzalo and Rätz, Stefanie and Rauterberg, Christoph and Redman, Bob and Redmer, Ronald and Reese, Daniel and Regibo, Sara and Reiners, Ansgar and Reinhold, Timo and Renie, Christian and Ribas, Ignasi and Ribeiro, Sergio and Ricciardi, Thiago Pereira and Rice, Ken and Richard, Olivier and Riello, Marco and Rieutord, Michel and Ripepi, Vincenzo and Rixon, Guy and Rockstein, Steve and Rodón Ortiz, José Ramón and Rodrigo Rodríguez, María Teresa and Rodríguez Amor, Alberto and Rodríguez Díaz, Luisa Fernanda and Rodriguez Garcia, Juan Pablo and Rodriguez-Gomez, Julio and Roehlly, Yannick and Roig, Fernando and Rojas-Ayala, Bárbara and Rolf, Tobias and Rørsted, Jakob Lysgaard and Rosado, Hugo and Rosotti, Giovanni and Roth, Olivier and Roth, Markus and Rousseau, Alex and Roxburgh, Ian and Roy, Fabrice and Royer, Pierre and Ruane, Kirk and Rufini Mastropasqua, Sergio and Ruiz De Galarreta, Claudia and Russi, Andrea and Saar, Steven and Saillenfest, Melaine and Salaris, Maurizio and Salmon, Sebastien and Saltas, Ippocratis and Samadi, Réza and Samadi, Aunia and Samra, Dominic and Sanches Da Silva, Tiago and Sánchez Carrasco, Miguel Andrés and Santerne, Alexandre and Santiago Pé, Amaia and Santoli, Francesco and Santos, Ängela R.G. and Sanz Mesa, Rosario and Sarro, Luis Manuel and Scandariato, Gaetano and Schäfer, Martin and Schlafly, Edward and Schmider, François Xavier and Schneider, Jean and Schou, Jesper and Schunker, Hannah and Schwarzkopf, Gabriel Jörg and Serenelli, Aldo and Seynaeve, Dries and Shan, Yutong and Shapiro, Alexander and Shipman, Russel and Sicilia, Daniela and Sierra Sanmartin, Maria Angeles and Sigot, Axelle and Silliman, Kyle and Silvotti, Roberto and Simon, Attila E. and Simoyama Napoli, Ricardo and Skarka, Marek and Smalley, Barry and Smiljanic, Rodolfo and Smit, Samuel and Smith, Alexis and Smith, Leigh and Snellen, Ignas and Sódor, Ádám and Sohl, Frank and Solanki, Sami K. and Sortino, Francesca and Sousa, Sérgio and Southworth, John and Souto, Diogo and Sozzetti, Alessandro and Stamatellos, Dimitris and Stassun, Keivan and Steller, Manfred and Stello, Dennis and Stelzer, Beate and Stiebeler, Ulrike and Stokholm, Amalie and Storelvmo, Trude and Strassmeier, Klaus and Strøm, Paul Anthony and Strugarek, Antoine and Sulis, Sophia and Švanda, Michal and Szabados, László and Szabó, Róbert and Szabó, Gyula M. and Szuszkiewicz, Ewa and Talens, Geert Jan and Teti, Daniele and Theisen, Tom and Thévenin, Frédéric and Thoul, Anne and Tiphene, Didier and Titz-Weider, Ruth and Tkachenko, Andrew and Tomecki, Daniel and Tonfat, Jorge and Tosi, Nicola and Trampedach, Regner and Traven, Gregor and Triaud, Amaury and Trønnes, Reidar and Tsantaki, Maria and Tschentscher, Matthias and Turin, Arnaud and Tvaruzka, Adam and Ulmer, Bernd and Ulmer-Moll, Solène and Ulusoy, Ceren and Umbriaco, Gabriele and Valencia, Diana and Valentini, Marica and Valio, Adriana and Valverde Guijarro, Ángel Luis and Van Eylen, Vincent and Van Grootel, Valerie and Van Kempen, Tim A. and Van Reeth, Timothy and Van Zelst, Iris and Vandenbussche, Bart and Vasiliou, Konstantinos and Vasilyev, Valeriy and Vaz De Mascarenhas, David and Vazan, Allona and Vela Nunez, Marina and Velloso, Eduardo Nunes and Ventura, Rita and Ventura, Paolo and Venturini, Julia and Vera Trallero, Isabel and Veras, Dimitri and Verdugo, Eva and Verma, Kuldeep and Vibert, Didier and Vicanek Martinez, Tobias and Vida, Krisztián and Vigan, Arthur and Villacorta, Antonio and Villaver, Eva and Villaverde Aparicio, Marcos and Viotto, Valentina and Vorobyov, Eduard and Vorontsov, Sergey and Wagner, Frank W. and Walton, Nicholas and Walton, Dave and Wang, Haiyang and Waters, Rens and Watson, Christopher and Wedemeyer, Sven and Weeks, Angharad and Weingrill, Jörg and Weiss, Annita and Wendler, Belinda and West, Richard and Westerdorff, Karsten and Westphal, Pierre Amaury and Wheatley, Peter and White, Tim and Whittaker, Amadou and Wickhusen, Kai and Wilson, Thomas and Windsor, James and Winter, Othon and Winther, Mark Lykke and Winton, Alistair and Witteck, Ulrike and Witzke, Veronika and Woitke, Peter and Wolter, David and Wuchterl, Günther and Wyatt, Mark and Yang, Dan and Yu, Jie and Zanmar Sanchez, Ricardo and Zapatero Osorio, María Rosa and Zechmeister, Mathias and Zhou, Yixiao and Ziemke, Claas and Zwintz, Konstanze and Böhm, Torsten and Dansac, Léo Michel}, issn = {1572-9508}, journal = {Experimental Astronomy}, number = {3}, publisher = {Springer Nature}, title = {{The PLATO mission}}, doi = {10.1007/s10686-025-09985-9}, volume = {59}, year = {2025}, } @article{19638, abstract = {The James Webb Space Telescope has revealed low-luminosity active galactic nuclei at redshifts of z ≳ 4–7, many of which host accreting massive black holes (BHs) with BH-to-galaxy mass (MBH/M⋆) ratios exceeding the local values by more than an order of magnitude. The origin of these overmassive BHs remains unclear but requires potential contributions from heavy seeds and/or episodes of super-Eddington accretion. We present a growth model coupled with dark matter halo assembly to explore the evolution of the MBH/M⋆ ratio under different seeding and feedback scenarios. Given the gas inflow rates in protogalaxies, BHs grow episodically at moderate super-Eddington rates, and the mass ratio increases early on, despite significant mass loss through feedback. Regardless of seeding mechanisms, the mass ratio converges to a universal value ∼0.1–0.3, set by the balance between gas feeding and star formation efficiency in the nucleus. This behavior defines an attractor in the MBH–M⋆ diagram, where overmassive BHs grow more slowly than their hosts, while undermassive seeds experience rapid growth before aligning with the attractor. We derive an analytical expression for the universal mass ratio, linking it to feedback strength and halo growth. The convergence of evolutionary tracks erases seeding information from the mass ratio by z ∼ 4–6. Detecting BHs with ∼105−6 M⊙ at higher redshifts that deviate from the convergence trend would provide key diagnostics of their birth conditions.}, author = {Hu, Haojie and Inayoshi, Kohei and Haiman, Zoltán and Ho, Luis C. and Ohsuga, Ken}, issn = {2041-8213}, journal = {The Astrophysical Journal Letters}, number = {2}, publisher = {IOP Publishing}, title = {{The convergence of heavy and light seeds to overmassive black holes at cosmic dawn}}, doi = {10.3847/2041-8213/adc680}, volume = {983}, year = {2025}, } @article{19640, abstract = {Synaptic plasticity is a key player in the brain’s life-long learning abilities. However, due to experimental limitations, the mechanistic link between synaptic plasticity rules and the network-level computations they enable remain opaque. Here we use evolutionary strategies (ES) to meta learn local co-active plasticity rules in large recurrent spiking networks with excitatory (E) and inhibitory (I) neurons, using parameterizations of increasing complexity. We discover rules that robustly stabilize network dynamics for all four synapse types acting in isolation (E-to-E, E-to-I, I-to-E and I-to-I). More complex functions such as familiarity detection can also be included in the search constraints. However, our meta learning strategy begins to fail for co-active rules of increasing complexity, as it is challenging to devise loss functions that effectively constrain network dynamics to plausible solutions a priori. Moreover, in line with previous work, we can find multiple degenerate solutions with identical network behaviour. As a local optimization strategy, ES provides one solution at a time and makes exploration of this degeneracy cumbersome. Regardless, we can glean the interdependecies of various plasticity parameters by considering the covariance matrix learned alongside the optimal rule with ES. Our work provides a proof of principle for the success of machine-learning-guided discovery of plasticity rules in large spiking networks, and points at the necessity of more elaborate search strategies going forward.}, author = {Confavreux, Basile J and Agnes, Everton J. and Zenke, Friedemann and Sprekeler, Henning and Vogels, Tim P}, issn = {1553-7358}, journal = {PLoS Computational Biology}, number = {4}, publisher = {Public Library of Science}, title = {{Balancing complexity, performance and plausibility to meta learn plasticity rules in recurrent spiking networks}}, doi = {10.1371/journal.pcbi.1012910}, volume = {21}, year = {2025}, } @article{19641, abstract = {Mycorrhizal and saprotrophic macromycetes contribute strongly to the carbon and nitrogen cycles of forest ecosystems, often studied by tracing stable isotope composition of carbon and nitrogen. The phenomenon of the saprotrophic-mycorrhizal divide highlights the difference in the stable isotope composition of fruiting bodies of mycorrhizal and saprotrophic fungi. Much less is known about the isotopic composition of the mycelium, which plays an important role in the formation of the soil organic matter and fuels the fungal trophic channel in soil food webs. In this study, we assessed whether the saprotrophic-mycorrhizal divide in the natural δ13С and δ15N values can be traced throughout entire fungal organisms. This hypothesis was tested using 16 species of ectomycorrhizal and six species of saprotrophic basidiomycetous fungi. We showed that not only fruiting bodies, but also the mycelium of ectomycorrhizal and saprotrophic fungi differs in the δ13C and δ15N values. In both ectomycorrhizal and saprotrophic fungi, the δ13C and δ15N values increased from mycelium to hymenophores and correlated positively with the total N content in the corresponding tissues. The differences between ectomycorrhizal and saprotrophic mycelium can be used to reconstruct the fungal-driven belowground carbon and nitrogen allocation, and the contribution of saprotrophic and mycorrhizal fungi to soil food webs.}, author = {Zuev, A. G. and Alexandrova, A. V. and Litvinskiy, V. A. and Pravdolyubova, Evgeniya and Tiunov, A. V.}, issn = {1432-1890}, journal = {Mycorrhiza}, number = {2}, publisher = {Springer Nature}, title = {{Saprotrophic-mycorrhizal divide in stable isotope composition throughout the whole fungus: From mycelium to hymenophore}}, doi = {10.1007/s00572-025-01203-w}, volume = {35}, year = {2025}, } @article{19642, abstract = {We study the criticality and subcriticality of powers (−Δ) α with α>0 of the discrete Laplacian −Δ acting on ℓ 2 (N). We prove that these positive powers of the Laplacian are critical if and only if α≥3/2. We complement our analysis with Hardy-type inequalities for (−Δ) α in the subcritical regimes α∈(0,3/2). As an illustration of the critical case α≥3/2, we analyze asymptotic properties of discrete eigenvalues emerging by coupling (−Δ) α with a localized potential.}, author = {Gerhát, Borbála M and Krejčiřík, David and Štampach, František}, issn = {2235-0616}, journal = {Revista Matematica Iberoamericana}, number = {3}, pages = {1173--1200}, publisher = {EMS Press}, title = {{Criticality transition for positive powers of the discrete Laplacian on the half line}}, doi = {10.4171/RMI/1523}, volume = {41}, year = {2025}, } @misc{19658, abstract = {We consider a family of totally asymmetric simple exclusion processes (TASEPs), consisting of particles on a lattice that require binding by a "token" in various physical configurations to advance over the lattice. Using a combination of theory and simulations, we address the following questions: (i) How token binding kinetics affects the current-density relation on the lattice; (ii) How this current-density relation depends on the scarcity of tokens; (iii) How tokens propagate the effects of the locally-imposed disorder (such as a slow site) over the entire lattice; (iv) How a shared pool of tokens couples concurrent TASEPs running on multiple lattices; (v) How our results translate to TASEPs with open boundaries that exchange particles with the reservoir. Since real particle motion (including in biological systems that inspired the standard TASEP model, e.g., protein synthesis or movement of molecular motors) is often catalyzed, regulated, actuated, or otherwise mediated, the token-driven TASEP dynamics analyzed in this paper should allow for a better understanding of real systems and enable a closer match between TASEP theory and experimental observations.}, author = {Tkačik, Gašper}, publisher = {Institute of Science and Technology Austria}, title = {{Token-driven totally asymmetric simple exclusion processes}}, doi = {10.15479/AT:ISTA:19658}, year = {2025}, } @article{19660, abstract = {We analyze the ground state energy of N fermions in a two-dimensional box interacting with an impurity particle via two-body point interactions. We show that for weak coupling, the ground state energy is asymptotically described by the polaron energy, as proposed by F. Chevy in the physics literature. The polaron energy is the solution of a nonlinear equation involving the Green’s function of the free Fermi gas and the binding energy of the two-body point interaction. We provide quantitative error estimates that are uniform in the thermodynamic limit.}, author = {Mitrouskas, David Johannes}, issn = {1432-0673}, journal = {Archive for Rational Mechanics and Analysis}, number = {3}, publisher = {Springer Nature}, title = {{The weakly coupled two-dimensional Fermi polaron}}, doi = {10.1007/s00205-025-02098-9}, volume = {249}, year = {2025}, } @article{19661, abstract = {The Nelson model describes non-relativistic particles coupled to a relativistic Bose scalar field. In this article, we study the renormalized version of the Nelson model with massless bosons in Davies' weak coupling limit. Our main result states that the two-body Coulomb potential emerges as an effective pair interaction between the particles, which arises from the exchange of virtual excitations of the quantum field.}, author = {Cárdenas, Esteban and Mitrouskas, David Johannes}, issn = {1751-8121}, journal = {Journal of Physics A: Mathematical and Theoretical}, number = {17}, publisher = {IOP Publishing}, title = {{The renormalized Nelson model in the weak coupling limit}}, doi = {10.1088/1751-8121/adcdd9}, volume = {58}, year = {2025}, } @article{19662, abstract = {We investigate the effect of changes in the Coriolis force caused by changes in the rotation rate on the top-of-atmosphere (TOA) radiant energy budget of an aquaplanet general circulation model with prescribed sea surface temperatures. We analyse the effective radiative forcing caused by changes from Earth-like rotation to values between 1/32 and 8 times the Earth's rotation rate. The forcing differs by about 60 W m−2 between the fastest and slowest rotation cases, with a monotonically increasing positive forcing for faster-than-Earth-like rotations and a non-monotonically increasing negative forcing for slower rotations. The largest contributions to the forcing are due to changes in, in this order, the shortwave cloud radiative effect (SWCRE) and the clear-sky outgoing longwave radiation (OLR). From the fastest to the slowest rotation, the Hadley cell expands and the troposphere becomes drier, increasing the OLR. This contributes to negative forcing at slower-than-Earth-like rotations and to positive forcing at faster-than-Earth-like rotations. The SWCRE is influenced by changes in the low-level cloudiness within the Hadley cell and the baroclinic regime. With the expansion of the Hadley cell, the area of enhanced tropospheric stability increases, resulting in more low-level clouds, a higher SWCRE, and increased negative forcing. The non-monotonicity results from an intermediate decrease in the SWCRE caused by the disappearance of baroclinic eddies as the Hadley cell reaches global extension. At rotations faster than Earth-like, the decrease in the SWCRE, mainly due to the weakening of baroclinic eddies and storm systems, leads to an increase in positive forcing. In summary, changes in the SWCRE, driven by different circulation responses at slower-than-Earth-like and faster-than-Earth-like rotations, strongly influence the TOA radiant energy budget. These effects, along with a substantial contribution from the clear-sky OLR, could impact the habitability of Earth-like rotating planets.}, author = {Gnanaraj, Abisha Mary and Bao, Jiawei and Schmidt, Hauke}, issn = {2698-4016}, journal = {Weather and Climate Dynamics}, number = {2}, pages = {489--503}, publisher = {Copernicus Publications}, title = {{The impact of the rotation rate on an aquaplanet's radiant energy budget: Insights from experiments varying the Coriolis parameter}}, doi = {10.5194/wcd-6-489-2025}, volume = {6}, year = {2025}, } @article{19663, abstract = {The centrosome is a microtubule orchestrator, nucleating and anchoring microtubules that grow radially and exert forces on cargos. At the same time, mechanical stresses from the microenvironment and cellular shape changes compress and bend microtubules. Yet, centrosomes are membraneless organelles, raising the question of how centrosomes withstand mechanical forces. Here, we discover that centrosomes can deform and even fracture. We reveal that centrosomes experience deformations during navigational pathfinding within motile cells. Coherence of the centrosome is maintained by Dyrk3 and cNAP1, preventing fracturing by forces. While cells can compensate for the depletion of centriolar-based centrosomes, the fracturing of centrosomes impedes cellular function by generating coexisting microtubule organizing centers that compete during path navigation and thereby cause cellular entanglement in the microenvironment. Our findings show that cells actively maintain the integrity of the centrosome to withstand mechanical forces. These results suggest that centrosome stability preservation is fundamental, given that almost all cells in multicellular organisms experience forces.}, author = {Schmitt, Madeleine T. and Kroll, Janina and Ruiz-Fernandez, Mauricio J.A. and Hauschild, Robert and Ghosh, Shaunak and Kameritsch, Petra and Merrin, Jack and Schmid, Johanna and Stefanowski, Kasia and Thomae, Andreas W. and Cheng, Jingyuan and Öztan, Gamze Naz and Konopka, Peter and Ortega, Germán Camargo and Penz, Thomas and Bach, Luisa and Baumjohann, Dirk and Bock, Christoph and Straub, Tobias and Meissner, Felix and Kiermaier, Eva and Renkawitz, Jörg}, issn = {2375-2548}, journal = {Science Advances}, number = {17}, publisher = {AAAS}, title = {{Protecting centrosomes from fracturing enables efficient cell navigation}}, doi = {10.1126/sciadv.adx4047}, volume = {11}, year = {2025}, } @article{19664, abstract = {Persistent revivals recently observed in Rydberg atom simulators have challenged our understanding of thermalization and attracted much interest to the concept of quantum many-body scars (QMBSs). QMBSs are non-thermal highly excited eigenstates that coexist with typical eigenstates in the spectrum of many-body Hamiltonians, and have since been reported in multiple theoretical models, including the so-called PXP model, approximately realized by Rydberg simulators. At the same time, questions of how common QMBSs are and in what models they are physically realized remain open. In this Letter, we demonstrate that QMBSs exist in a broader family of models that includes and generalizes PXP to longer-range constraints and states with different periodicity. We show that in each model, multiple QMBS families can be found. Each of them relies on a different approximate algebra, leading to oscillatory dynamics in all cases. However, in contrast to the PXP model, their observation requires launching dynamics from weakly entangled initial states rather than from a product state. QMBSs reported here may be experimentally probed using Rydberg atom simulator in the regime of longer-range Rydberg blockades.}, author = {Kerschbaumer, Aron and Ljubotina, Marko and Serbyn, Maksym and Desaules, Jean-Yves Marc}, issn = {1079-7114}, journal = {Physical Review Letters}, number = {16}, publisher = {American Physical Society}, title = {{Quantum many-body scars beyond the PXP model in Rydberg simulators}}, doi = {10.1103/PhysRevLett.134.160401}, volume = {134}, year = {2025}, } @inproceedings{19665, abstract = {As AI-based decision-makers increasingly influence human lives, it is a growing concern that their decisions may be unfair or biased with respect to people's protected attributes, such as gender and race. Most existing bias prevention measures provide probabilistic fairness guarantees in the long run, and it is possible that the decisions are biased on any decision sequence of fixed length. We introduce *fairness shielding*, where a symbolic decision-maker---the fairness shield---continuously monitors the sequence of decisions of another deployed black-box decision-maker, and makes interventions so that a given fairness criterion is met while the total intervention costs are minimized. We present four different algorithms for computing fairness shields, among which one guarantees fairness over fixed horizons, and three guarantee fairness periodically after fixed intervals. Given a distribution over future decisions and their intervention costs, our algorithms solve different instances of bounded-horizon optimal control problems with different levels of computational costs and optimality guarantees. Our empirical evaluation demonstrates the effectiveness of these shields in ensuring fairness while maintaining cost efficiency across various scenarios.}, author = {Cano Cordoba, Filip and Henzinger, Thomas A and Könighofer, Bettina and Kueffner, Konstantin and Mallik, Kaushik}, booktitle = {Proceedings of the 39th AAAI Conference on Artificial Intelligence}, issn = {2374-3468}, location = {Philadelphia, PA, United States}, number = {15}, pages = {15659--15668}, publisher = {Association for the Advancement of Artificial Intelligence}, title = {{Fairness shields: Safeguarding against biased decision makers}}, doi = {10.1609/aaai.v39i15.33719}, volume = {39}, year = {2025}, } @inproceedings{19666, abstract = {Markov decision processes (MDP) are a well-established model for sequential decision-making in the presence of probabilities. In *robust* MDP (RMDP), every action is associated with an *uncertainty set* of probability distributions, modelling that transition probabilities are not known precisely. Based on the known theoretical connection to stochastic games, we provide a framework for solving RMDPs that is generic, reliable, and efficient. It is *generic* both with respect to the model, allowing for a wide range of uncertainty sets, including but not limited to intervals, L1- or L2-balls, and polytopes; and with respect to the objective, including long-run average reward, undiscounted total reward, and stochastic shortest path. It is *reliable*, as our approach not only converges in the limit, but provides precision guarantees at any time during the computation. It is *efficient* because -- in contrast to state-of-the-art approaches -- it avoids explicitly constructing the underlying stochastic game. Consequently, our prototype implementation outperforms existing tools by several orders of magnitude and can solve RMDPs with a million states in under a minute.}, author = {Meggendorfer, Tobias and Weininger, Maximilian and Wienhöft, Patrick}, booktitle = {Proceedings of the 39th AAAI Conference on Artificial Intelligence}, issn = {2374-3468}, location = {Philadelphia, PA, United States}, number = {25}, pages = {26631--26641}, publisher = {Association for the Advancement of Artificial Intelligence}, title = {{Solving robust Markov decision processes: Generic, reliable, efficient}}, doi = {10.1609/aaai.v39i25.34865}, volume = {39}, year = {2025}, } @inproceedings{19667, abstract = {The problem of checking satisfiability of linear real arithmetic (LRA) and non-linear real arithmetic (NRA) formulas has broad applications, in particular, they are at the heart of logic-related applications such as logic for artificial intelligence, program analysis, etc. While there has been much work on checking satisfiability of unquantified LRA and NRA formulas, the problem of checking satisfiability of quantified LRA and NRA formulas remains a significant challenge. The main bottleneck in the existing methods is a computationally expensive quantifier elimination step. In this work, we propose a novel method for efficient quantifier elimination in quantified LRA and NRA formulas. We propose a template-based Skolemization approach, where we automatically synthesize linear/polynomial Skolem functions in order to eliminate quantifiers in the formula. The key technical ingredient in our approach are Positivstellensätze theorems from algebraic geometry, which allow for an efficient manipulation of polynomial inequalities. Our method offers a range of appealing theoretical properties combined with a strong practical performance. On the theory side, our method is sound, semi-complete, and runs in subexponential time and polynomial space, as opposed to existing sound and complete quantifier elimination methods that run in doubly-exponential time and at least exponential space. On the practical side, our experiments show superior performance compared to state of the art SMT solvers in terms of the number of solved instances and runtime, both on LRA and on NRA benchmarks.}, author = {Chatterjee, Krishnendu and Kafshdar Goharshadi, Ehsan and Karrabi, Mehrdad and Motwani, Harshit J. and Seeliger, Maximilian and Zikelic, Dorde}, booktitle = {Proceedings of the 39th AAAI Conference on Artificial Intelligence}, issn = {2374-3468}, location = {Philadelphia, PA, United States}, number = {11}, pages = {11158--11166}, publisher = {Association for the Advancement of Artificial Intelligence}, title = {{Quantified linear and polynomial arithmetic satisfiability via template-based skolemization}}, doi = {10.1609/aaai.v39i11.33213}, volume = {39}, year = {2025}, } @inproceedings{19668, abstract = {Learning-based methods provide a promising approach to solving highly non-linear control tasks that are often challenging for classical control methods. To ensure the satisfaction of a safety property, learning-based methods jointly learn a control policy together with a certificate function for the property. Popular examples include barrier functions for safety and Lyapunov functions for asymptotic stability. While there has been significant progress on learning-based control with certificate functions in the white-box setting, where the correctness of the certificate function can be formally verified, there has been little work on ensuring their reliability in the black-box setting where the system dynamics are unknown. In this work, we consider the problems of certifying and repairing neural network control policies and certificate functions in the black-box setting. We propose a novel framework that utilizes runtime monitoring to detect system behaviors that violate the property of interest under some initially trained neural network policy and certificate. These violating behaviors are used to extract new training data, that is used to re-train the neural network policy and the certificate function and to ultimately repair them. We demonstrate the effectiveness of our approach empirically by using it to repair and to boost the safety rate of neural network policies learned by a state-of-the-art method for learning-based control on two autonomous system control tasks.}, author = {Yu, Zhengqi and Zikelic, Dorde and Henzinger, Thomas A}, booktitle = {Proceedings of the 39th AAAI Conference on Artificial Intelligence}, issn = {2374-3468}, location = {Philadelphia, PA, United States}, number = {25}, pages = {26409--26417}, publisher = {Association for the Advancement of Artificial Intelligence}, title = {{Neural control and certificate repair via runtime monitoring}}, doi = {10.1609/aaai.v39i25.34840}, volume = {39}, year = {2025}, } @inproceedings{19669, abstract = {We consider a class of optimization problems defined by a system of linear equations with min and max operators. This class of optimization problems has been studied under restrictive conditions, such as, (C1) the halting or stability condition; (C2) the non-negative coefficients condition; (C3) the sum upto 1 condition; and (C4) the only min or only max operator condition. Several seminal results in the literature focus on special cases. For example, turn-based stochastic games correspond to conditions C2 and C3; and Markov decision process to conditions C2, C3, and C4. However, the systematic computational complexity study of all the cases has not been explored, which we address in this work. Some highlights of our results are: with conditions C2 and C4, and with conditions C3 and C4, the problem is NP-complete, whereas with condition C1 only, the problem is in UP intersects coUP. Finally, we establish the computational complexity of the decision problem of checking the respective conditions.}, author = {Chatterjee, Krishnendu and Luo, Ruichen and Saona Urmeneta, Raimundo J and Svoboda, Jakub}, booktitle = {Proceedings of the 39th AAAI Conference on Artificial Intelligence}, issn = {2374-3468}, location = {Philadelphia, PA, United States}, number = {11}, pages = {11150--11157}, publisher = {Association for the Advancement of Artificial Intelligence}, title = {{Linear equations with min and max operators: Computational complexity}}, doi = {10.1609/aaai.v39i11.33212}, volume = {39}, year = {2025}, }