Reliable Data-Driven Decision Making in Cyber-Physical Systems

Log Barriers for Safe Non-convex Black-box Optimization

Autores: Usmanova, Ilnura; Krause, Andreas; Kamgarpour, Maryam
Publicado en: arXiv, Edición 22, 2022
Editor: arXiv
DOI: 10.48550/arxiv.1912.09478

Noise Regularization for Conditional Density Estimation

Autores: Rothfuss, Jonas; Ferreira, Fabio; Boehm, Simon; Walther, Simon; Ulrich, Maxim; Asfour, Tamim; Krause, Andreas
Publicado en: arXiv, Edición 5, 2019
Editor: arXiv

Bayesian Optimisation for Fast and Safe Parameter Tuning of SwissFEL

Autores: Kirschner, Johannes; Adelmann, Andreas; Hiller, Nicole; Ischebeck, Rasmus; Krause, Andreas; Mutný, Mojmir; Nonnenmacher, Manuel
Publicado en: FEL2019, Proceedings of the 39th International Free-Electron Laser Conference, Edición 6, 2019
Editor: FEL Conference
DOI: 10.3929/ethz-b-000385955

Information Directed Sampling for Linear Partial Monitoring

Autores: Kirschner, Johannes; Lattimore, Tor; Krause, Andreas
Publicado en: Proc. International Conference on Learning Theory (COLT), 2020
Editor: PMLR

Interactively Learning Preference Constraints in Linear Bandits

Autores: David Lindner; Tschiatschek, Sebastian; Hofmann, Katja; Krause, Andreas
Publicado en: Proceedings of Machine Learning Research, 162, Edición 8, 2022, Página(s) 13505-13527
Editor: PMLR
DOI: 10.48550/arxiv.2206.05255

Safe Exploration for Interactive Machine Learning

Autores: Turchetta, Matteo; Berkenkamp, Felix; Krause, Andreas
Publicado en: Advances in Neural Information Processing Systems (NeurIPS), Edición 17, 2019
Editor: Advances in Neural Information Processing Systems

Contextual Games: Multi-Agent Learning with Side Information

Autores: Sessa, Pier Giuseppe; id_orcid0000-0001-8986-8815; Bogunovic, Ilija; Krause, Andreas; Kamgarpour, Maryam; id_orcid0000-0003-0230-3518
Publicado en: Advances in Neural Information Processing Systems (NeurIPS), Edición 17, 2020
Editor: NeurIPS Foundation

Safe non-smooth black-box optimization with application to policy search

Autores: Usmanova, Ilnura; Krause, Andreas; Kamgarpour, Maryam
Publicado en: Proc. Conference on Learning for Dynamics and Control (L4DC), Edición 5, 2020
Editor: PMLR

Experimental Design for Optimization of Orthogonal Projection Pursuit Models

Autores: Mojmir Mutny; Johannes Kirschner; Andreas Krause
Publicado en: Proc. AAAI Conference on Artificial Intelligence, Edición 7, 2020
Editor: AAAI
DOI: 10.1609/aaai.v34i06.6585

Adaptive Gaussian Process Change Point Detection

Autores: Caldarelli, Edoardo; Wenk, Philippe; Bauer, Stefan; Krause, Andreas
Publicado en: Proceedings of Machine Learning Research, 162, Edición 16, 2022, Página(s) 2542-2571
Editor: PMLR

Active Bayesian Causal Inference

Autores: Toth, Christian; Lorch, Lars; Knoll, Christian; Krause, Andreas; Pernkopf, Franz; Peharz, Robert; von Kügelgen, Julius
Publicado en: Advances in Neural Information Processing Systems 35, Edición 22, 2022, Página(s) 16261 - 16275
Editor: Curran
DOI: 10.48550/arxiv.2206.02063

Safe and Efficient Model-free Adaptive Control via Bayesian Optimization

Autores: Christopher König; Matteo Turchetta; John Lygeros; Alisa Rupenyan; Andreas Krause
Publicado en: International Conference on Robotics and Automation, Edición 2, 2021
Editor: IEEE
DOI: 10.1109/icra48506.2021.9561349

Meta-Learning Priors for Safe Bayesian Optimization

Autores: Rothfuss, Jonas; Koenig, Christopher; Rupenyan-Vasileva, Alisa Bohos; Krause, Andreas
Publicado en: Proc. Conference on Robot Learning (CoRL) 205, Edición 24, 2022, Página(s) 237-265
Editor: PMLR
DOI: 10.48550/arxiv.2210.00762

Safe Model-Based Multi-Agent Mean-Field Reinforcement Learning

Autores: Matej Jusup, Barna Pásztor, Tadeusz Janik, Kenan Zhang, Francesco Corman, Andreas Krause, Ilija Bogunovic
Publicado en: Proc. of the 23rd International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2024), 2024
Editor: IFAAMAS

Stochastic Linear Bandits Robust to Adversarial Attacks

Autores: Bogunovic, Ilija; Losalka, Arpan; Krause, Andreas; Scarlett, Jonathan
Publicado en: Proc. Conference on Artificial Intelligence and Statistics (AISTATS), Edición 10, 2021
Editor: PMLR

Meta-Learning Hypothesis Spaces for Sequential Decision-making

Autores: Kassraie, Parnian; Rothfuss, Jonas; Krause, Andreas
Publicado en: Proceedings of Machine Learning Research, 162, Edición 33, 2022, Página(s) 10802 - 10824
Editor: PMLR
DOI: 10.48550/arxiv.2202.00602

Structured Variational Inference in Partially Observable Unstable Gaussian Process State Space Models

Autores: Curi, Sebastian; Melchior, Silvan; Berkenkamp, Felix; id_orcid0000-0002-5179-6606; Krause, Andreas
Publicado en: Proc. Conference on Learning for Dynamics and Control (L4DC), Edición 2, 2020
Editor: PMLR

Safe Contextual Bayesian Optimization for Sustainable Room Temperature PID Control Tuning

Autores: Marcello Fiducioso, Sebastian Curi, Benedikt Schumacher, Markus Gwerder, Andreas Krause
Publicado en: Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence, 2019, Página(s) 5850-5856, ISBN 978-0-9992411-4-1
Editor: International Joint Conferences on Artificial Intelligence Organization
DOI: 10.24963/ijcai.2019/811

Distributional Gradient Matching for Learning Uncertain Neural Dynamics Models

Autores: Treven, Lenart; Wenk, Philippe; Dörfler, Florian; Krause, Andreas
Publicado en: Neural Information Processing Systems (NeurIPS), Edición 14, 2021
Editor: Advances in Neural Information Processing Systems
DOI: 10.3929/ethz-b-000521449

Amortized Inference for Causal Structure Learning

Autores: Lorch, Lars; id_orcid0000-0001-7465-5892; Sussex, Scott; Rothfuss, Jonas; Krause, Andreas; Schölkopf, Bernhard
Publicado en: Advances in Neural Information Processing Systems 35, Edición 41, 2022, Página(s) 13104-13118
Editor: Curran
DOI: 10.3929/ethz-b-000589141

DiBS: Differentiable Bayesian Structure Learning

Autores: Lorch, Lars; Rothfuss, Jonas; Schölkopf, Bernhard; Krause, Andreas
Publicado en: Proc. Neural Information Processing Systems (NeurIPS), Edición 9, 2021
Editor: Advances in Neural Information Processing Systems

Safe Reinforcement Learning via Confidence-Based Filters

Autores: Curi, Sebastian; Lederer, Armin; Hirche, Sandra; Krause, Andreas
Publicado en: IEEE Conference on Decision and Control (CDC), Edición 41, 2022, Página(s) 3409 - 3415
Editor: IEEE
DOI: 10.1109/cdc51059.2022.9992470

Distributionally Robust Bayesian Optimization

Autores: Kirschner, Johannes; Bogunovic, Ilija; Jegelka, Stefanie; Krause, Andreas
Publicado en: Proc. Conference on Artificial Intelligence and Statistics (AISTATS), Edición 20, 2020
Editor: PMLR

Lifelong Bandit Optimization: No Prior and No Regret

Autores: Schur, Felix; Kassraie, Parnian; Rothfuss, Jonas; Krause, Andreas
Publicado en: Proceedings of the Thirty-Ninth Conference on Uncertainty in Artificial Intelligence 216, 2023, Página(s) 1847-1857
Editor: PMLR

ODIN: ODE-Informed Regression for Parameter and State Inference in Time-Continuous Dynamical Systems

Autores: Philippe Wenk; Gabriele Abbati; Michael A. Osborne; Bernhard Schölkopf; Andreas Krause; Stefan Bauer
Publicado en: AAAI Conference on Artificial Intelligence, Edición 2, 2020
Editor: AAAI
DOI: 10.1609/aaai.v34i04.6106

The Dynamics of Riemannian Robbins-Monro Algorithms

Autores: Karimi, Mohammad Reza; Hsieh, Ya-Ping; Mertikopoulos, Panayotis; Krause, Andreas
Publicado en: COLT 2022 - 35th Annual Conference on Learning Theory / 178, Edición 10, 2022, Página(s) 3503
Editor: PMLR
DOI: 10.3929/ethz-b-000587230

Hallucinated Adversarial Control for Conservative Offline Policy Evaluation

Autores: Rothfuss, Jonas; Sukhija, Bhavya; Birchler, Tobias; Kassraie, Parnian; Krause, Andreas
Publicado en: Proceedings of the Thirty-Ninth Conference on Uncertainty in Artificial Intelligence, Edición 41, 2023, Página(s) 1774-1784
Editor: PMLR
DOI: 10.48550/arxiv.2303.01076

Efficient Model-Based Reinforcement Learning through Optimistic Policy Search and Planning

Autores: Curi, Sebastian; Berkenkamp, Felix; id_orcid0000-0002-5179-6606; Krause, Andreas
Publicado en: Advances in Neural Information Processing Systems (NeurIPS), Edición 8, 2020
Editor: NeurIPS Foundation

Active Exploration for Inverse Reinforcement Learning

Autores: David Lindner; Krause, Andreas; Ramponi, Giorgia
Publicado en: Proc. Neural Information Processing Systems (NeurIPS) 35, Edición 17, 2022, ISBN 9781713871088
Editor: Curran
DOI: 10.48550/arxiv.2207.08645

Risk-averse Heteroscedastic Bayesian Optimization

Autores: Anastasiia Makarova, Ilnura Usmanova, Ilija Bogunovic, Andreas Krause
Publicado en: Proc. Neural Information Processing Systems (NeurIPS), 2021
Editor: Advances in Neural Information Processing Systems

Causal Modeling with Stationary Diffusions

Autores: Lorch, Lars; Krause, Andreas; Schölkopf, Bernhard
Publicado en: Proceedings of the 27th International Conference on Artifi- cial Intelligence and Statistics (AISTATS) 2024, Edición 72, 2024
Editor: Proceedings of Machine Learning Research
DOI: 10.48550/arxiv.2310.17405

Adaptive Sampling for Stochastic Risk-Averse Learning

Autores: Curi, Sebastian; Levy, Kfir. Y.; Jegelka, Stefanie; Krause, Andreas
Publicado en: Advances in Neural Information Processing Systems (NeurIPS), Edición 8, 2020
Editor: NeurIPS Foundation

Stochastic Bandits with Context Distributions

Autores: Kirschner, Johannes; Krause, Andreas
Publicado en: Advances in Neural Information Processing Systems (NeurIPS), Edición 21, 2019
Editor: Advances in Neural Processing Systems
DOI: 10.3929/ethz-b-000385952

Learning Safety Constraints from Demonstrations with Unknown Rewards

Autores: David Lindner, Xin Chen, Sebastian Tschiatschek, Katja Hofmann, Andreas Krause
Publicado en: Proceedings of the 27th International Conference on Artifi- cial Intelligence and Statistics (AISTATS) 2024, 2024
Editor: Proceedings of Machine Learning Research

Meta-Learning Reliable Priors in the Function Space

Autores: Rothfuss, Jonas; Heyn, Dominique; Chen, Jinfan; Krause, Andreas
Publicado en: Proc. Neural Information Processing Systems (NeurIPS), Edición 15, 2021
Editor: Advances in Neural Information Processing Systems

Efficient Pure Exploration for Combinatorial Bandits with Semi-Bandit Feedback

Autores: Jourdan, Marc; Mutný, Mojmír; Kirschner, Johannes; Krause, Andreas
Publicado en: International Conference on Algorithmic Learning Theory, Edición 1, 2021
Editor: Proceedings of Machine Learning Research

Graph Neural Network Bandits

Autores: Kassraie, Parnian; Krause, Andreas; Bogunovic, Ilija
Publicado en: Advances in Neural Information Processing Systems 35, Edición 16, 2022, Página(s) 34519-34531
Editor: Curran
DOI: 10.48550/arxiv.2207.06456

Efficient Model-based Multi-agent Reinforcement Learning via Optimistic Equilibrium Computation

Autores: Sessa, Pier Giuseppe; Kamgarpour, Maryam; Krause, Andreas
Publicado en: Proceedings of Machine Learning Research, 162, Edición 12, 2022, Página(s) 19580-19597
Editor: PMLR
DOI: 10.3929/ethz-b-000591032

Misspecified Gaussian Process Bandit Optimization

Autores: Ilija Bogunovic, Andreas Krause
Publicado en: Proc. Neural Information Processing Systems (NeurIPS), 2021
Editor: Advances in Neural Information Processing Systems

No-Regret Learning in Unknown Games with Correlated Payoffs

Autores: Sessa, Pier Giuseppe; Bogunovic, Ilija; Kamgarpour, Maryam; Krause, Andreas
Publicado en: Advances in Neural Information Processing Systems (NeurIPS), Edición 2, 2019
Editor: Advances in Neural Processing Systems
DOI: 10.3929/ethz-b-000383403

Projection Free Online Learning Over Smooth Sets

Autores: Levy, Kfir Y.; Krause, Andreas
Publicado en: Proc. Conference on Artificial Intelligence and Statistics (AISTATS), Edición 5, 2019
Editor: PMLR

PACOH: Bayes-Optimal Meta-Learning with PAC-Guarantees

Autores: Rothfuss, Jonas; Fortuin, Vincent; Josifoski, Martin; Krause, Andreas
Publicado en: Proc. International Conference on Machine Learning (ICML), Edición 2, 2021
Editor: PMLR

Adversarial Causal Bayesian Optimization

Autores: Scott Sussex, Pier Giuseppe Sessa, Anastasiia Makarova, Andreas Krause
Publicado en: Proc. International Conference on Learning Representations (ICLR), 2024
Editor: OpenReview

Logistic Q-Learning

Autores: Bas-Serrano, Joan; Curi, Sebastian; Krause, Andreas; Neu, Gergely
Publicado en: Proceedings of the 24th International Conference on Artificial Intelligence and Statistics (AISTATS 2021), Edición 3, 2021
Editor: Proceedings of Machine Learning Research

MARS: Meta-Learning as Score Matching in the Function Space

Autores: Pavasovic, Krunoslav Lehman; Rothfuss, Jonas; Krause, Andreas
Publicado en: International Conference on Learning Representations (ICLR), Edición 37, 2023
Editor: OpenReview
DOI: 10.48550/arxiv.2210.13319

Risk-Averse Offline Reinforcement Learning

Autores: Urpí, Núria Armengol; Curi, Sebastian; Krause, Andreas
Publicado en: Proc. International Conference on Learning Representations (ICLR), Edición 10, 2021
Editor: openreview

Neural Contextual Bandits without Regret

Autores: Kassraie, Parnian; Krause, Andreas
Publicado en: International Conference on Artificial Intelligence and Statistics (AISTATS), Edición 8, 2022
Editor: Proceedings of Machine Learning Research

Model-based Causal Bayesian Optimization

Autores: Sussex, Scott; Makarova, Anastasia; Krause, Andreas
Publicado en: International Conference on Learning Representations (ICLR), Edición 10, 2023
Editor: OpenReview

Log Barriers for Safe Non-convex Black-box Optimization

Autores: Usmanova, Ilnura; Krause, Andreas; Kamgarpour, Maryam
Publicado en: Proc. Conference on Learning for Dynamics and Control (L4DC), Edición 6, 2020
Editor: PMLR

Adaptive and Safe Bayesian Optimization in High Dimensions via One-Dimensional Subspaces

Autores: Kirschner, Johannes; Mutný, Mojmir; Hiller, Nicole; Ischebeck, Rasmus; Krause, Andreas
Publicado en: Proc. International Conference on Machine Learning (ICML), Edición 22, 2019
Editor: PMLR
DOI: 10.3929/ethz-b-000385951

Submodular Reinforcement Learning

Autores: Prajapat, Manish; Mutný, Mojmír; Zeilinger, Melanie N.; Krause, Andreas
Publicado en: Proc. International Conference on Learning Representations (ICLR), Edición 70, 2024
Editor: OpenReview
DOI: 10.48550/arxiv.2307.13372

Combining Pessimism with Optimism for Robust and Efficient Model-Based Deep Reinforcement Learning

Autores: Curi, Sebastian; Bogunovic, Ilija; Krause, Andreas
Publicado en: Proc. International Conference on Machine Learning (ICML), Edición 2, 2021
Editor: PMLR

Learning to Play Sequential Games versus Unknown Opponents

Autores: Sessa, Pier Giuseppe; id_orcid0000-0001-8986-8815; Bogunovic, Ilija; Kamgarpour, Maryam; id_orcid0000-0003-0230-3518; Krause, Andreas
Publicado en: Advances in Neural Information Processing Systems (NeurIPS), Edición 13, 2020
Editor: NeurIPS Foundation

Safe Convex Learning under Uncertain Constraints

Autores: Usmanova, Ilnura; Krause, Andreas; Kamgarpour, Maryam
Publicado en: Proc. Conference on Artificial Intelligence and Statistics (AISTATS), Edición 16, 2019
Editor: PMLR
DOI: 10.3929/ethz-b-000386455

Mixed Strategies for Robust Optimization of Unknown Objectives

Autores: Sessa, Pier Giuseppe; Bogunovic, Ilija; Kamgarpour, Maryam; Krause, Andreas
Publicado en: Proc. Conference on Artificial Intelligence and Statistics (AISTATS), Edición 6, 2019
Editor: PMLR

Online Submodular Resource Allocation with Applications to Rebalancing Shared Mobility Systems

Autores: Sessa, Pier Giuseppe; id_orcid0000-0001-8986-8815; Bogunovic, Ilija; Krause, Andreas; Kamgarpour, Maryam; id_orcid0000-0003-0230-3518
Publicado en: International Conference on Machine Learning, Edición 2, 2021
Editor: Proceedings of Machine Learning Research, 139
DOI: 10.3929/ethz-b-000522280

Learning Stable Deep Dynamics Models for Partially Observed or Delayed Dynamical Systems

Autores: Schlaginhaufen, Andreas; Wenk, Philippe; Krause, Andreas; Dörfler, Florian; id_orcid0000-0002-9649-5305
Publicado en: Advances in Neural Information Processing Systems 34 pre-proceedings (NeurIPS 2021), Edición 7, 2021
Editor: Advances in Neural Information Processes

BaCaDI: Bayesian Causal Discovery with Unknown Interventions

Autores: Hägele, Alexander; Rothfuss, Jonas; Lorch, Lars; id_orcid0000-0001-7465-5892; Somnath, Vignesh Ram; Schölkopf, Bernhard; Krause, Andreas
Publicado en: Proc. International Conference on Artificial Intelligence and Statistics (AISTATS), Edición 19, 2023
Editor: PMLR
DOI: 10.48550/arxiv.2206.01665

Bounding Inefficiency of Equilibria in Continuous Actions Games using Submodularity and Curvature

Autores: Sessa, Pier Giuseppe; Kamgarpour, Maryam; Krause, Andreas
Publicado en: Proc. Conference on Artificial Intelligence and Statistics (AISTATS), Edición 7, 2019
Editor: PMLR
DOI: 10.3929/ethz-b-000383405

The Schrödinger Bridge between Gaussian Measures has a Closed Form

Autores: Bunne, Charlotte, Hsieh, Ya-Ping, Cuturi, Marco, Krause, Andreas
Publicado en: International Conference on Artificial Intelligence and Statistics (AISTATS), 2023
Editor: PMLR

Movement Penalized Bayesian Optimization with Application to Wind Energy Systems

Autores: Ramesh, Shyam Sundhar; Sessa, Pier Giuseppe; Krause, Andreas; Bogunovic, Ilija
Publicado en: Advances in Neural Information Processing Systems 35, Edición 16, 2022, Página(s) 27036 - 27048, ISBN 978-1-7138-7108-8
Editor: Curran
DOI: 10.48550/arxiv.2210.08087

Corruption-Tolerant Gaussian Process Bandit Optimization

Autores: Bogunovic, Ilija; Krause, Andreas; Scarlett, Jonathan
Publicado en: Proc. Conference on Artificial Intelligence and Statistics (AISTATS), Edición 1, 2020
Editor: PMLR

Bias-Robust Bayesian Optimization via Dueling Bandits

Autores: Kirschner, Johannes; Krause, Andreas
Publicado en: International Conference on Machine Learning, Edición 3, 2021
Editor: Proceedings of Machine Learning Research, 139

Safe Reinforcement Learning via Curriculum Induction

Autores: Turchetta, Matteo; Kolobov, Andrey; Shah, Shital; Krause, Andreas; Agarwal, Alekh
Publicado en: Advances in Neural Information Processing Systems (NeurIPS), Edición 1, 2020
Editor: NeurIPS Foundation

Epistemic Uncertainty for Practical Deep Model-Based Reinforcement Learning

Autores: Curi, Sebastian
Publicado en: Edición 5, 2022
Editor: ETH Zurich
DOI: 10.3929/ethz-b-000579085

Multi-Player Bandits: The Adversarial Case

Autores: Alatur, Pragnya; Levy, Kfir Y.; Krause, Andreas
Publicado en: Journal of Machine Learning Research, Edición 4, 2020, ISSN 1533-7928
Editor: Journal of Machine Learning Research
DOI: 10.3929/ethz-b-000414972

No-Regret Bayesian Optimization with Unknown Hyperparameters

Autores: Berkenkamp, Felix; Schoellig, Angela P.; Krause, Andreas
Publicado en: Journal of Machine Learning Research, 2019, ISSN 1533-7928
Editor: Journal of Machine Learning Research

Tuning particle accelerators with safety constraints using Bayesian optimization

Autores: Johannes Kirschner; Mojmir Mutný; Andreas Krause; Jaime Coello de Portugal; Nicole Hiller; Jochem Snuverink
Publicado en: Physical Review Accelerators and Beams, 25 (6), Edición 37, 2022, Página(s) 062802, ISSN 2469-9888
Editor: American Physical Society
DOI: 10.1103/physrevaccelbeams.25.062802

GOSAFEOPT : Scalable safe exploration for global optimization of dynamical systems

Autores: Bhavya Sukhija; Matteo Turchetta; David Lindner; Andreas Krause; Sebastian Trimpe; Dominik Baumann
Publicado en: Artificial Intelligence, 320, Edición 2, 2023, ISSN 0004-3702
Editor: Elsevier BV
DOI: 10.1016/j.artint.2023.103922

Efficient Model-Based Multi-Agent Mean-Field Reinforcement Learning

Autores: Pásztor, Barna; Krause, Andreas; Bogunovic, Ilija
Publicado en: Transactions on Machine Learning Research, Edición 10, 2023, ISSN 2835-8856
Editor: TMLR
DOI: 10.48550/arxiv.2107.04050

Scalable PAC-Bayesian Meta-Learning via the PAC-Optimal Hyper-Posterior: From Theory to Practice

Autores: Jonas Rothfuss, Martin Josifoski, Vincent Fortuin, Andreas Krause
Publicado en: Journal of Machine Learning Research (JMLR), Edición 24 (386), 2023, Página(s) 1-62, ISSN 1532-4435
Editor: MIT Press

Publicaciones

Compartir esta página Compartir esta página en las redes sociales

Descargar Descargar el contenido de la página