Thomas Schön
Professor i artificiell intelligens vid Institutionen för informationsteknologi; Systemteknik
- Telefon:
- 018-471 25 94
- Mobiltelefon:
- 073-593 38 87
- E-post:
- thomas.schon@it.uu.se
- Besöksadress:
- Hus 10, Lägerhyddsvägen 1
- Postadress:
- Box 337
751 05 UPPSALA
- Akademiska meriter:
- Docent
Mer information visas för dig som medarbetare om du loggar in.
Kort presentation
Thomas B. Schön är Beijerprofessor i artificiell intelligens vid institutionen för informationsteknologi vid Uppsala universitet och vetenskaplig ledare för det femåriga projektet AI4Research. Han doktorerade i reglerteknik i februari 2006 vid Linköpings universitet, och har tidigare varit verksam vid University of Cambridge (Storbritannien), University of Newcastle (Australien) och Técnica Federico Santa María (Valparaíso, Chile).
Nyckelord
- ai4research
- artificial intelligence
- artificiell intelligens
- automatic control
- computer vision
- data analytics
- datorseende
- decision-making with algorithms
- deep learning
- machine learning
- maskininlärning
- reglerteknik
- signal processing
- signalbehandling
Biografi
Thomas B. Schön är Beijerprofessor i artificiell intelligens vid institutionen för informationsteknologi vid Uppsala universitet och vetenskaplig ledare för det femåriga projektet AI4Research. Han doktorerade i reglerteknik i februari 2006, tog sin Masterexamen i tillämpad fysik och elektroteknik i september 2001 och sin kandidatexamen i företagsekonomi 2001, samtliga vid Linköpings universitet. Han har tidigare varit verksam vid University of Cambridge (Storbritannien), University of Newcastle (Australien) och Técnica Federico Santa María (Valparaíso, Chile). 2018 valdes han in i Kungl. Ingenjörsvetenskapsakademien och Kungliga Vetenskaps-Societeten (Uppsala). Han belönades med Tage Erlanders pris i naturvetenskap och teknik 2017 och Arnbergska priset 2016, som båda utdelas av Kungl. Vetenskapsakademien. Han belönades år 2014 med Automatica Best Paper Prize och 2013 med Best PhD Thesis Award av the European Association for Signal Processing 2009 utsågs han till bästa lärare vid Linköpings universitets tekniska högskola. Han är medlem i IEEE (Institute of Electrical and Electronics Engineers) och i ELLIS Society.
Thomas Schön har ett brett intresse för att utveckla nya algoritmer och matematiska modeller som kan lära sig och agera utifrån data. Hans huvudsakliga vetenskapliga fält är maskininlärning, men han publicerar regelbundet inom andra fält såsom statistik, reglerteknik, signalbehandling och datorseende. Han bedriver både grundforskning och tillämpad forskning, där det sistnämnda oftast genomförs i samarbete med industri eller tillämpade forskargrupper.
Publikationer
Senaste publikationer
- Safe Reinforcement Learning in Uncertain Contexts (2024)
- Machine Learning Based Prediction of Imminent ICP Insults During Neurocritical Care of Traumatic Brain Injury (2024)
- Uncertainty Estimation with Recursive Feature Machines (2024)
- Online learning in motion modeling for intra-interventional image sequences (2024)
- Rao-Blackwellized particle smoothing for simultaneous localization and mapping (2024)
Alla publikationer
Artiklar
- Safe Reinforcement Learning in Uncertain Contexts (2024)
- Machine Learning Based Prediction of Imminent ICP Insults During Neurocritical Care of Traumatic Brain Injury (2024)
- Rao-Blackwellized particle smoothing for simultaneous localization and mapping (2024)
- On the Equivalence of Direct and Indirect Data-Driven Predictive Control Approaches (2024)
- On the trade-off between event-based and periodic state estimation under bandwidth constraints (2023)
- Variational Elliptical Processes (2023)
- Variational system identification for nonlinear state-space models (2023)
- Invertible Kernel PCA With Random Fourier Features (2023)
- Smoothed State Estimation via Efficient Solution of Linear Equations (2023)
- Diffusion-Based 3D Motion Estimation from Sparse 2D Observations (2023)
- How Reliable is Your Regression Model’s Uncertainty Under Real-World Distribution Shifts? (2023)
- Inferring the probability distribution over strain tensors in polycrystals from diffraction based measurements (2023)
- Overparameterized Linear Regression Under Adversarial Attacks (2023)
- Screening for Chagas disease from the electrocardiogram using a deep neural network (2023)
- Guarantees for data-driven control of nonlinear systems using semidefinite programming (2023)
- On the regularization in DeePC (2023)
- Online Learning for Prediction via Covariance Fitting (2023)
- Sequential Monte Carlo (2023)
- Neural motion planning in dynamic environments (2023)
- Probabilistic Estimation of Instantaneous Frequencies of Chirp Signals (2023)
- Development and validation of deep learning ECG-based prediction of myocardial infarction in emergency department patients (2022)
- Data to Controller for Nonlinear Systems (2022)
- Memory efficient constrained optimization of scanning-beam lithography (2022)
- Direct Transmittance Estimation in Heterogeneous Participating Media Using Approximated Taylor Expansions (2022)
- Predicting Political Violence Using a State-Space Model (2022)
- Incorporating Sum Constraints into Multitask Gaussian Processes (2022)
- Efficient Learning of the Parameters of Non-Linear Models Using Differentiable Resampling in Particle Filters (2022)
- Nonlinear System Identification (2022)
- Quantifying the Uncertainty of the Relative Geometry in Inertial Sensors Arrays (2021)
- Gaussian Variational State Estimation for Nonlinear State-Space Models (2021)
- Deep neural network-estimated electrocardiographic age as a mortality predictor (2021)
- Universal probabilistic programming offers a powerful approach to statistical phylogenetics (2021)
- Stochastic quasi-Newton with line-search regularisation (2021)
- Learning Robust LQ-Controllers Using Application Oriented Exploration (2020)
- On the smoothness of nonlinear system identification (2020)
- Smoothing With Couplings of Conditional Particle Filters (2020)
- Automatic diagnosis of the 12-lead ECG using a deep neural network (2020)
- The effect of interventions on COVID-19 (2020)
- Machine Learning in Risk Prediction (2020)
- Nonlinear Input Design as Optimal Control of a Hamiltonian System (2020)
- Optimal controller/observer gains of discounted-cost LQG systems (2019)
- Getting started with particle Metropolis-Hastings for inference in nonlinear dynamical models (2019)
- Neutron transmission strain tomography for non-constant stress-free lattice spacing (2019)
- Deep kernel learning for integral measurements (2019)
- A Fast and Robust Algorithm for Orientation Estimation Using Inertial Sensors (2019)
- Data consistency approach to model validation (2019)
- High-Dimensional Filtering Using Nested Sequential Monte Carlo (2019)
- Elements of Sequential Monte Carlo (2019)
- Probabilistic approach to limited-data computed tomography reconstruction (2019)
- Identification of a Duffing oscillator using particle Gibbs with ancestor sampling (2019)
- Data Consistency Approach to Model Validation (2019)
- On model order priors for Bayesian identification of SISO linear systems (2019)
- Probabilistic modelling and reconstruction of strain (2018)
- Automated learning with a probabilistic programming language: Birch (2018)
- Probabilistic learning of nonlinear dynamical systems using sequential Monte Carlo (2018)
- Modeling and Interpolation of the Ambient Magnetic Field by Gaussian Processes (2018)
- Learning of state-space models with highly informative observations (2018)
- Maximum likelihood identification of stable linear dynamical systems (2018)
- System identification through online sparse Gaussian process regression with input noise (2017)
- Smoothed State Estimation via Efficient Solution of Linear Equations (2017)
- Divide-and-Conquer with sequential Monte Carlo (2017)
- A flexible state–space model for learning nonlinear dynamical systems (2017)
- On robust input design for nonlinear dynamical models (2017)
- Mean and variance of the LQG cost function (2016)
- Using convolution to estimate the score function for intractable state-transition models (2016)
- Magnetometer calibration using inertial sensors (2016)
- Rao–Blackwellized particle smoothers for conditionally linear Gaussian models (2016)
- Particle Metropolis–Hastings using gradient and Hessian information (2015)
- On the exponential convergence of the Kaczmarz algorithm (2015)
- A new structure exploiting derivation of recursive direct weight optimization (2015)
- Indoor positioning using ultrawideband and inertial measurements (2015)
- Particle Gibbs with ancestor sampling (2014)
- Automated segmentation of synchrotron-scanned fossils
- No Double Descent in Principal Component Regression: A High-Dimensional Analysis
- Latent linear dynamics in spatiotemporal medical data
- Registration by tracking for sequential 2D MRI
- ECG-Based Electrolyte Prediction: Evaluating Regression and Probabilistic Methods
- Derivation and validation of STOPSMOKE: An instrument built from Swedish population data for predicting smoking abstinence post myocardial infarction
Böcker
Konferenser
- Uncertainty Estimation with Recursive Feature Machines (2024)
- Online learning in motion modeling for intra-interventional image sequences (2024)
- On Feynman–Kac training of partial Bayesian neural networks (2024)
- NTIRE 2023 HR NonHomogeneous Dehazing Challenge Report (2023)
- Lens-to-Lens Bokeh Effect Transformation (2023)
- Regularization properties of adversarially-trained linear regression (2023)
- Refusion (2023)
- Gaussian inference for data-driven state-feedback design of nonlinear systems (2023)
- NTIRE 2023 Image Shadow Removal Challenge Report (2023)
- NTIRE 2023 Challenge on Stereo Image Super-Resolution (2023)
- Unsupervised dynamic modeling of medical image transformations (2022)
- Learning Proposals for Practical Energy-Based Regression (2022)
- Learning deep autoregressive models for hierarchical data (2021)
- Variational State and Parameter Estimation (2021)
- Willems' fundamental lemma based on second-order moments (2021)
- Bayes Control of Hammerstein Systems (2021)
- First Steps Towards Self-Supervised Pretraining of the 12-Lead ECG (2021)
- Deep State Space Models for Nonlinear System Identification (2021)
- Learning a Deformable Registration Pyramid (2021)
- Accurate 3D Object Detection using Energy-Based Models (2021)
- Deep Energy-Based NARX Models (2021)
- How convolutional neural networks deal with aliasing (2021)
- Beyond Occam's Razor in System Identification (2021)
- Energy-Based Models for Deep Probabilistic Regression (2020)
- Evaluating Scalable Bayesian Deep Learning Methods for Robust Computer Vision (2020)
- How to Train Your Energy-Based Model for Regression (2020)
- Probabilistic programming for birth-death models of evolution using an alive particle filter with delayed sampling (2020)
- Particle Filter with Rejection Control and Unbiased Estimator of the Marginal Likelihood (2020)
- Deep Learning and System Identification (2020)
- Beyond exploding and vanishing gradients (2020)
- Automatic 12-lead ECG Classification Using a Convolutional Network Ensemble (2020)
- A fast quasi-Newton-type method for large-scale stochastic optimisation (2020)
- Deep convolutional networks in system identification (2019)
- Inferring Heterogeneous Causal Effects in Presence of Spatial Confounding (2019)
- Conditionally Independent Multiresolution Gaussian Processes (2019)
- Robust exploration in linear quadratic reinforcement learning (2019)
- Bayesian identification of state-space models via adaptive thermostats (2019)
- Evaluating model calibration in classification (2019)
- Data-driven impulse response regularization via deep learning (2018)
- Auxiliary-Particle-Filter-based Two-Filter Smoothing for Wiener State-Space Models (2018)
- Delayed sampling and automatic Rao-Blackwellization of probabilistic programs (2018)
- Learning localized spatio-temporal models from streaming data (2018)
- Automatic diagnosis of short-duration 12-lead ECG using a deep convolutional network (2018)
- Probabilistic programming allows for automated inference in factor graph models (2018)
- Learning nonlinear state-space models using smooth particle-filter-based likelihood approximations (2018)
- How consistent is my model with the data? (2018)
- Learning convex bounds for linear quadratic control policy synthesis (2018)
- Regularized parametric system identification (2018)
- Linearly constrained Gaussian processes (2017)
- On the construction of probabilistic Newton-type algorithms (2017)
- Prediction Performance After Learning in Gaussian Process Regression (2017)
- A scalable and distributed solution to the inertial motion capture problem (2016)
- Accelerometer calibration using sensor fusion with a gyroscope (2016)
- Computationally Efficient Bayesian Learning of Gaussian Process State Space Models (2016)
- Particle-based Gaussian process optimization for input design in nonlinear dynamical models (2016)
- Prediction performance after learning in Gaussian process regression (2016)
- Particle filtering based identification for autonomous nonlinear ODE models (2015)
- Marginalizing Gaussian process hyperparameters using sequential Monte Carlo (2015)
- Nonlinear state space smoothing using the conditional particle filter (2015)
- Nonlinear state space model identification using a regularized basis function expansion (2015)
- On Identification via EM with Latent Disturbances and Lagrangian Relaxation (2015)
- Bayesian nonparametric identification of piecewise affine ARX systems (2015)
- Robust auxiliary particle filters using multiple importance sampling (2014)
- Backward sequential Monte Carlo for marginal smoothing (2014)
- Identification of jump Markov linear models using particle filters (2014)
Rapporter
