Card Overview

Stochastics Meeting Lunteren 2023

Stochastics Meeting Lunteren 2023 51th annual meeting of the Dutch probability and statistics community. 13 November 2023 10:30 till 15 November 2023 14:30 - Location: De Werelt, Westhofflaan 2, 6741 KH Lunteren | Add to my calendar Speakers Topics Programme Organizing Comittee Participation fee and lodging Registration History Marek Biskup (UCLA Los Angeles) Alexandra Carpentier (Potsdam) Alessandra Cipriani (London) Mathias Drton (Munich) Eviatar Procaccia (Technion Haifa) Aaditya Ramdas (Carnegie Mellon) Marek Biskup (UCLA Los Angeles) ( https://www.math.ucla.edu/~biskup/ ) 1. Extreme order statistics of logarithmically correlated processes, Part 1 2. Extreme order statistics of logarithmically correlated processes, Part 2 Alexandra Carpentier (Potsdam) ( https://sites.google.com/site/alexandracarpentierresearch/ ) 1. Ranking problems in crowd-sourcing, Part 1 2. Ranking problems in crowd-sourcing, Part 2 Alessandra Cipriani (London) ( https://sites.google.com/site/aciprian41/ ) 1. Random interfaces and sandpiles Part I 2. Random interfaces and sandpiles Part II Mathias Drton (Munich) ( https://www.math.cit.tum.de/math/personen/professuren/drton-mathias/ ) 1: Consistent Tests of Independence via Rank Statistics 2: Identification and Estimation of Graphical Continuous Lyapunov Models Eviatar Procaccia (Technion Haifa) ( https://sites.google.com/site/ebprocaccia/ ) 1. Vertex-removal stability and the least positive value of harmonic measures. 2. Fluctuations of Stationary Hastings Levitov are log-correlated. Aaditya Ramdas (Carnegie Mellon) ( https://www.stat.cmu.edu/~aramdas/ ) 1. Game-theoretic testing: e-values, martingales and betting 2. Game-theoretic estimation: confidence sequences Poster titles Juntong Chen: Estimating a regression function in exponential families by model selection Gabriel Clara: Dropout Regularization Versus $\ell_2$-Penalization in the Linear Model Alexander Dürre: Multivariate Lp Location Testing: Wald Tests and Lagrange Multipler Tests based on Simplices Francesco Gili: Adaptive and Efficient Isotonic Estimation in Wicksell’s Problem Yanyan Hu: Large deviation for diffusion processes with finite state-dependent fast switching on complete Riemannian manifold Aljosa Marjanovic: TBA Riccardo Michielan: Percolation phase transition on geometric intersection graphs Rounak Ray: Percolation on Preferential Attachment Model Monday 10:30-11:15 Coffee & Registration 11:15-11:30 Opening 11:30-12:30 Ramdas 12:45 Lunch 15:30-16:00 Tea 16:00-17:00 Ramdas 17:00-18:00 Biskup 18:45 Dinner Vergadering Poster Session Tuesday 08:00-09:00 Breakfast 09:00-10:00 Biskup 10:00-10:30 Coffee 10:30-11:30 Carpentier 11:30-12:30 Cipriani 12:45 Lunch 15:30-16:00 Tea 16:00-17:00 Carpentier 17:00-18:00 Drton 18:45 Dinner Wednesday 07:30-08:30 Breakfast 08:30-09:30 Cipriani 09:30-10:30 Procaccia 10:30-11:00 Coffee 11:00-12:00 Procaccia 12:00-13:00 Drton 13:15 Lunch R Kraaij (Delft) W. Th. F. den Hollander (Leiden) A. van der Vaart (Delft) A limited number of people may stay at the conference centre overnight. Rooms are assigned on a first come, first serve, basis. The following prices are valid for those who registered before October 1st: Package Price Full stay (registration fee, accomodation and all meals from Monday befor lunch to Wednesday after lunch) €510,- Two day arrangement (registration fee, one overnight stay and all meals during two days Monday and Tuesday €340,- Tuesday and Wednesday €295,- Participation in the scientific programme without overnight stay, per day Monday or Tuesday €140,- Wednesday €110,- Everyone who wishes to participate in the meeting, including those who will not stay overnight nor take meals, is requested to register before October 1st. Register here Please refer to this list to see all previous speakers. Travel De Werelt is easily reached by train. The train station is Lunteren . See the train schedule or the public transport schedule . From the train station it is a 15-20 minute walk to De Werelt. See the path on Open Street Maps or Google maps . Directions for travel by car can be found on the web site of De Werelt . The forests surrounding De Werelt show all colours in fall and offer nice hiking opportunities during the afternoon. Bring solid shoes with you, though, and watch out for the weather on the rain radar .

Profile of a Safety & Security Researcher: Riccardo Ferrari

Somewhat unexpectedly, Riccardo Ferrari ’s choice of a niche topic for his PhD, fault diagnosis, led to an exciting second life for his field of research. “You normally use fault diagnosis to monitor physical degradation and faults, but we extended it to deal with cyber threats. With one theory, we can now make systems like wind turbines or aeroplanes resilient to faults and secure industrial plants or autonomous vehicles against cyber attacks. Our next step? Teach an AI how devices, like electric car batteries, age. This is the kind of exciting cross-disciplinary research we want to do with colleagues at the TU Delft Safety & Security Institute.” Following a PhD in Information Engineering from the University of Trieste, Riccardo spent seven years in industrial R&D in the steelmaking sector. He specialised in instrumentation and automation solutions for process control, first as a researcher and then as executive manager. His career path is considered unusual due to the return to academics after a comfortable position in industry. How did he get there? “My first choice for further education was architecture, but I dropped the idea when my mother told me that my drawing skills were not good enough! My decision to study engineering was an element of chance. I used to spend time at my father’s car mechanic shop after school, watching him work. Maybe it sparked curiosity for engineering.” “When I went on to a Masters in Electronic Engineering at the University of Trieste, I did an industrial thesis about detecting defects inside a steel object. We figured out that if you apply tension to that object and pluck it, like when tuning a guitar string, its sound can reveal if it is faulty! So acoustics and musical intuition helped to solve the problem.” Riccardo’s interest in acoustics did not stop there. Alongside his MSc, Riccardo also juggled a BA in Classical Piano. “Yes, well, I almost got kicked out of the Conservatory. I wasn’t getting the grades because I was simultaneously working on my engineering thesis. Luckily, I pulled it off and graduated (Grade 109/110). It was hard, but I believed it wasn’t impossible. My (perhaps naive) view was that I could get a result if I put in the energy and effort. I also had the opportunity to write another thesis on acoustics! This time I added an engineering sparkle to a musical problem and built a computer model to analyse the influence of a pianist’s touch on the sound you get out of the instrument.” When the same company that supported his Master’s thesis offered Riccardo a PhD followed by an engineer position, it was an easy and logical choice. “I transitioned from academic to corporate life smoothly. But as the amount of research declined so did my interest. It was the path with the least resistance. In retrospect, I wonder whether it was the best path.” “During that period, I was still into research but had to write scientific papers during my free time. This double life became untenable when our first daughter was born, for good reasons! I reevaluated my motivations and decided to quit my comfortable job and start a research position at Delft. All based on a one-year postdoc contract. Some might call it madness. But again, I believed that effort, energy and a bit of good luck would get me there.” Riccardo Ferrari is an Associate Professor in Fault Tolerant Control at the Delft Center for Systems and Control, within the Faculty of Mechanical, Maritime, and Materials Engineering (3mE) at TU Delft. Riccardo investigates how to make dynamical systems resilient against faults, malicious cyber-attacks and degradation phenomena whilst simultaneously fighting uncertainty. text Helen Hartmann illustration De Zagerij Download article

Filter results

Stochastics Meeting Lunteren 2023

Stochastics Meeting Lunteren 2023 51th annual meeting of the Dutch probability and statistics community. 13 November 2023 10:30 till 15 November 2023 14:30 - Location: De Werelt, Westhofflaan 2, 6741 KH Lunteren | Add to my calendar Speakers Topics Programme Organizing Comittee Participation fee and lodging Registration History Marek Biskup (UCLA Los Angeles) Alexandra Carpentier (Potsdam) Alessandra Cipriani (London) Mathias Drton (Munich) Eviatar Procaccia (Technion Haifa) Aaditya Ramdas (Carnegie Mellon) Marek Biskup (UCLA Los Angeles) ( https://www.math.ucla.edu/~biskup/ ) 1. Extreme order statistics of logarithmically correlated processes, Part 1 2. Extreme order statistics of logarithmically correlated processes, Part 2 Alexandra Carpentier (Potsdam) ( https://sites.google.com/site/alexandracarpentierresearch/ ) 1. Ranking problems in crowd-sourcing, Part 1 2. Ranking problems in crowd-sourcing, Part 2 Alessandra Cipriani (London) ( https://sites.google.com/site/aciprian41/ ) 1. Random interfaces and sandpiles Part I 2. Random interfaces and sandpiles Part II Mathias Drton (Munich) ( https://www.math.cit.tum.de/math/personen/professuren/drton-mathias/ ) 1: Consistent Tests of Independence via Rank Statistics 2: Identification and Estimation of Graphical Continuous Lyapunov Models Eviatar Procaccia (Technion Haifa) ( https://sites.google.com/site/ebprocaccia/ ) 1. Vertex-removal stability and the least positive value of harmonic measures. 2. Fluctuations of Stationary Hastings Levitov are log-correlated. Aaditya Ramdas (Carnegie Mellon) ( https://www.stat.cmu.edu/~aramdas/ ) 1. Game-theoretic testing: e-values, martingales and betting 2. Game-theoretic estimation: confidence sequences Poster titles Juntong Chen: Estimating a regression function in exponential families by model selection Gabriel Clara: Dropout Regularization Versus $\ell_2$-Penalization in the Linear Model Alexander Dürre: Multivariate Lp Location Testing: Wald Tests and Lagrange Multipler Tests based on Simplices Francesco Gili: Adaptive and Efficient Isotonic Estimation in Wicksell’s Problem Yanyan Hu: Large deviation for diffusion processes with finite state-dependent fast switching on complete Riemannian manifold Aljosa Marjanovic: TBA Riccardo Michielan: Percolation phase transition on geometric intersection graphs Rounak Ray: Percolation on Preferential Attachment Model Monday 10:30-11:15 Coffee & Registration 11:15-11:30 Opening 11:30-12:30 Ramdas 12:45 Lunch 15:30-16:00 Tea 16:00-17:00 Ramdas 17:00-18:00 Biskup 18:45 Dinner Vergadering Poster Session Tuesday 08:00-09:00 Breakfast 09:00-10:00 Biskup 10:00-10:30 Coffee 10:30-11:30 Carpentier 11:30-12:30 Cipriani 12:45 Lunch 15:30-16:00 Tea 16:00-17:00 Carpentier 17:00-18:00 Drton 18:45 Dinner Wednesday 07:30-08:30 Breakfast 08:30-09:30 Cipriani 09:30-10:30 Procaccia 10:30-11:00 Coffee 11:00-12:00 Procaccia 12:00-13:00 Drton 13:15 Lunch R Kraaij (Delft) W. Th. F. den Hollander (Leiden) A. van der Vaart (Delft) A limited number of people may stay at the conference centre overnight. Rooms are assigned on a first come, first serve, basis. The following prices are valid for those who registered before October 1st: Package Price Full stay (registration fee, accomodation and all meals from Monday befor lunch to Wednesday after lunch) €510,- Two day arrangement (registration fee, one overnight stay and all meals during two days Monday and Tuesday €340,- Tuesday and Wednesday €295,- Participation in the scientific programme without overnight stay, per day Monday or Tuesday €140,- Wednesday €110,- Everyone who wishes to participate in the meeting, including those who will not stay overnight nor take meals, is requested to register before October 1st. Register here Please refer to this list to see all previous speakers. Travel De Werelt is easily reached by train. The train station is Lunteren . See the train schedule or the public transport schedule . From the train station it is a 15-20 minute walk to De Werelt. See the path on Open Street Maps or Google maps . Directions for travel by car can be found on the web site of De Werelt . The forests surrounding De Werelt show all colours in fall and offer nice hiking opportunities during the afternoon. Bring solid shoes with you, though, and watch out for the weather on the rain radar .

Profile of a Safety & Security Researcher: Riccardo Ferrari

Somewhat unexpectedly, Riccardo Ferrari ’s choice of a niche topic for his PhD, fault diagnosis, led to an exciting second life for his field of research. “You normally use fault diagnosis to monitor physical degradation and faults, but we extended it to deal with cyber threats. With one theory, we can now make systems like wind turbines or aeroplanes resilient to faults and secure industrial plants or autonomous vehicles against cyber attacks. Our next step? Teach an AI how devices, like electric car batteries, age. This is the kind of exciting cross-disciplinary research we want to do with colleagues at the TU Delft Safety & Security Institute.” Following a PhD in Information Engineering from the University of Trieste, Riccardo spent seven years in industrial R&D in the steelmaking sector. He specialised in instrumentation and automation solutions for process control, first as a researcher and then as executive manager. His career path is considered unusual due to the return to academics after a comfortable position in industry. How did he get there? “My first choice for further education was architecture, but I dropped the idea when my mother told me that my drawing skills were not good enough! My decision to study engineering was an element of chance. I used to spend time at my father’s car mechanic shop after school, watching him work. Maybe it sparked curiosity for engineering.” “When I went on to a Masters in Electronic Engineering at the University of Trieste, I did an industrial thesis about detecting defects inside a steel object. We figured out that if you apply tension to that object and pluck it, like when tuning a guitar string, its sound can reveal if it is faulty! So acoustics and musical intuition helped to solve the problem.” Riccardo’s interest in acoustics did not stop there. Alongside his MSc, Riccardo also juggled a BA in Classical Piano. “Yes, well, I almost got kicked out of the Conservatory. I wasn’t getting the grades because I was simultaneously working on my engineering thesis. Luckily, I pulled it off and graduated (Grade 109/110). It was hard, but I believed it wasn’t impossible. My (perhaps naive) view was that I could get a result if I put in the energy and effort. I also had the opportunity to write another thesis on acoustics! This time I added an engineering sparkle to a musical problem and built a computer model to analyse the influence of a pianist’s touch on the sound you get out of the instrument.” When the same company that supported his Master’s thesis offered Riccardo a PhD followed by an engineer position, it was an easy and logical choice. “I transitioned from academic to corporate life smoothly. But as the amount of research declined so did my interest. It was the path with the least resistance. In retrospect, I wonder whether it was the best path.” “During that period, I was still into research but had to write scientific papers during my free time. This double life became untenable when our first daughter was born, for good reasons! I reevaluated my motivations and decided to quit my comfortable job and start a research position at Delft. All based on a one-year postdoc contract. Some might call it madness. But again, I believed that effort, energy and a bit of good luck would get me there.” Riccardo Ferrari is an Associate Professor in Fault Tolerant Control at the Delft Center for Systems and Control, within the Faculty of Mechanical, Maritime, and Materials Engineering (3mE) at TU Delft. Riccardo investigates how to make dynamical systems resilient against faults, malicious cyber-attacks and degradation phenomena whilst simultaneously fighting uncertainty. text Helen Hartmann illustration De Zagerij Download article
48212 results

Half height card - Default

Styling based on the availability of image, title, metadata and text

Tracing ancient settlements in Colombia with remote sensing

A team of the LDE alliance (Leiden University, TU Delft, and Erasmus University Rotterdam) asked whether it might be possible to search for signs of ancient settlements in the jungle with affordable remote sensing techniques. For an expedition in a Colombian dense forest, the team, including remote sensing expert Felix Dahle of TU Delft, joined forces with archaeologists and drone experts from Colombia. In mountainous forests, drones provide affordable access to areas that would otherwise be unreachable from the ground. A LiDAR laser scanner already proved its value in coastal observation . The big question was whether LiDAR could bypass the many treetops. Trees reflect the laser, so it was crucial to fly close so it found its way through the foliage. The team mounted a highly portable LiDAR laser scanner to a drone and went on expedition nearby ancient terraces of the Tairona culture in the Sierra Nevada of Santa Marta. “We had to find the sweet spot. Close to the archaeological sites and still secure above the canopy”, says Felix Dahle. And it passed the test. The LiDAR laser scanner create a point cloud and a detailed 3D model of the landscape. “We were able to detect ancient terraces in the jungle. We discovered that we can scan through the forest when it is not too dense, but some areas remained unfathomable. We could also distinguish several types of vegetation, which might be of great use too to find undiscovered archaeological sites.”

Full card - image & title only

No results matching your search query were found.

Full card - half image, title

No results matching your search query were found.

Full card - half image, title and abstract

No results matching your search query were found.