A
Alberto Dalla Libera
Researcher at University of Padua
Publications - 25
Citations - 76
Alberto Dalla Libera is an academic researcher from University of Padua. The author has contributed to research in topics: Kernel (statistics) & Computer science. The author has an hindex of 3, co-authored 18 publications receiving 34 citations. Previous affiliations of Alberto Dalla Libera include Mitsubishi Electric Research Laboratories.
Papers
More filters
Journal ArticleDOI
Model-Based Reinforcement Learning for Physical Systems Without Velocity and Acceleration Measurements
TL;DR: In this article, a derivative-free model learning framework for reinforcement learning algorithms based on Gaussian Process Regression (GPR) is proposed, where the state is defined as the set of past position measurements, instead of representing the system state as suggested by the physics with a collection of positions, velocities and accelerations.
Journal ArticleDOI
A novel Multiplicative Polynomial Kernel for Volterra series identification
TL;DR: A new regularization network for Volterra models identification relies on a new kernel given by the product of basic building blocks, which allows to better select the monomials that really influence the system output, much increasing the prediction capability of the model.
Journal ArticleDOI
Autonomous Learning of the Robot Kinematic Model
TL;DR: An algorithm is presented that learns a forward kinematics model of a robot starting from a time series of visual observations and relies on a Gaussian process (GP) model based on a polynomial kernel.
Posted Content
A novel Multiplicative Polynomial Kernel for Volterra series identification
TL;DR: In this paper, a new regularization network for nonlinear system identification is proposed, which relies on a new kernel given by the product of basic building blocks, each block contains some unknown parameters that can be estimated from data using marginal likelihood optimization.
Journal ArticleDOI
A Data-Efficient Geometrically Inspired Polynomial Kernel for Robot Inverse Dynamic
TL;DR: In this paper, a data-driven inverse dynamics estimator based on Gaussian Process Regression (GIP) is proposed, which is based on the recently introduced Multiplicative Polynomial Kernel (MPK), a redefinition of the classical polynomial kernel equipped with a set of parameters that allows for a higher regularization.