Subscribe to our newsletter

Events

Courses

Information Theory in Biology, Phd Programme on Data Science and Computation

University of Bologna

Seminars

DateSpeakerTitle
23 June, 2021
Markus Vincze - TU Wien, Austria
Robot Object Detection and Grasping
Live Stream
25 May, 2021
Gabriele Steidl - TU Berlin, Germany
TBA
Live Stream
11 May, 2021
Domenico Marinucci - University of Rome “Tor Vergata”, Italy
TBA
Live Stream
27 April, 2021
Elena Nicora - University of Genoa, Italy
TBA
Live Stream
13 April, 2021
Francesco Orabona - Boston University, USA
TBA
Live Stream
16 March, 2021
Luca Garello - University of Genoa and IIT, Italy
Towards Imitation Learning in Robotics
Live Stream
16 March, 2021
Giulia Luise - Imperial College London, UK
Regularity properties of Entropic Optimal Transport in applications to machine learning
Live Stream

Markus Vincze

TU Wien, Austria

Abstract
In the near future robots will operate next to humans. Robots will be expected to know about all the objects in the domain where they are working. This will require methods to rapidly learn new objects, recognise and manipulate them. The talk will present recent advances of learning objects from CAD models, learning models for more robust object pose estimation, and the application in a tidy up scenario.

Bio
Markus Vincze is Professor at TU Wien leading the Vision for Robotics (V4R) team with the objective to make robots see. Markus received his diploma in mechanical engineering from TU Wien in 1988, M.Sc. from Rensselaer Polytechnic Institute, USA, 1990, and PhD at TU Wien 1993. As PostDoc he worked at HelpMate Robotics Inc. with Josef Engelberger and the Vision Laboratory of Gregory Hager at Yale University. 2004 he obtained the habilitation in robotics based on the work in visual servoing. He coordinated EC projects RobVision, ActIPret, robotshome and Hobbit and contributed to many other such as GRASP, STRANDS, and Squirrel. He was the program chair of ICRA 2013 in Karlsruhe and organized ERF 2015 and HRI 2017 in Vienna. He is director on the board of euRobotics. Markus' special interests are cognitive vision methods for robotics solutions situated in real-world environments and especially homes.

Gabriele Steidl

TU Berlin, Germany

Abstract
TBA

Bio
TBA

Domenico Marinucci

University of Rome “Tor Vergata”, Italy

Abstract
TBA

Bio
TBA

Elena Nicora

University of Genoa, Italy

Abstract
TBA

Bio
TBA

Francesco Orabona

Boston University

Abstract
TBA

Bio
TBA

Francesco Orabona

Boston University

Abstract
TBA

Bio
TBA

Luca Garello

University of Genoa and IIT

Abstract
Imitation learning plays a key role in our development from the early years of our life. In fact, by observing expert demonstrators we are able to learn new skills. For this reason the idea of having robots able to learn new tasks by using demonstration policies is the subject of an increasing number of research. With our work we focus on the ability of remapping actions in our perspective and we propose a generative model able to shift the perspective from third person to first person. This perspective translation is performed by using only RGB images. Moreover, our model generates an embedded representation of the action which can be used to understand the action. These embeddings are autonomously learnt following a time-consistent pattern without the human supervision. In the last part of the seminar we will show how our model can be successfully implemented on a real robot to perform an imitation task.

Bio
Luca is a 2nd year PhD student at MALGA and collaborates with the Robotics Brain and Cognitive Sciences laboratory at the Italian Institute of Technology (iit). His research interests revolve around machine learning and computer vision applied to Robotics, with a focus on new algorithms that enhance the quality of Human-Robot-Interaction.

Giulia Luise

UCL London, UK

Abstract
The entropic regularization has proved to be a powerful tool to define approximations of optimal transport distances with improved computational and statistical aspects. In this talk we will focus on further advantages of such entropic regularization, in terms of smoothness. We discuss its regularity properties and their role in some machine learning problems where regularized optimal transport is used as discrepancy metric in supervised and unsupervised frameworks.

Bio
Giulia Luise has recently obtained her PhD in Machine Learning at UCL, London, under the supervision of Massimiliano Pontil and Carlo Ciliberto. Her main research interest focuses on the interplay of optimal transport and machine learning. She is now a Research Associate at Imperial College, where she started working on reinforcement learning.

Workshops

Foundations of Algorithmic Fairness

12, 16 March - 2021

As machine learning becomes more and more spread in our society, the potential risk of using algorithms that behave unfairly is rising. As a result there is growing interest to design learning methods that meet fairness requirements. Several ELLIS units across Europe decided to come together and organize a 2 days workshop on the topic of algorithmic fairness. The aim of this event is to provide a platform for presenting recent advances in this area, with a focus on algorithms foundations, as well as discussing future research opportunities. During these two days our invited speakers are going to present the latest developments covering a variety of learning problems.

More Info Stream Day 1 Stream Day 2
© Ellis Unit Genova
Subscribe