68 research outputs found
Less is more: Nyström computational regularization
No full textWe study Nyström type subsampling approaches to large scale kernel methods, and prove learning bounds in the statistical learning setting, where random sampling and high probability estimates are considered. In particular, we prove that these approaches can achieve optimal learning bounds, provided the subsampling level is suitably chosen. These results suggest a simple incremental variant of Nyström kernel ridge regression, where the subsampling level controls at the same time regularization and computations. Extensive experimental analysis shows that the considered approach achieves state of the art performances on benchmark large scale datasets
Generalization properties and implicit regularization for multiple passes SGM
No full textWe study the generalization properties of stochastic gradient methods for learning with convex loss functions and linearly parameterized functions. We show that, in the absence of penalizations or constraints, the stability and approximation properties of the algorithm can be controlled by tuning either the step-size or the number of passes over the data. In this view, these parameters can be seen to control a form of implicit regularization. Numerical results complement the theoretical findings
Dataset for: "Incremental Semiparametric Inverse Dynamics Learning"
No full text<p>Dataset used in the experimental section of the paper:</p>
<blockquote>
<p>R. Camoriano, S. Traversaro, L. Rosasco, G. Metta and F. Nori, "<strong>Incremental semiparametric inverse dynamics learning,</strong>" <em>2016 IEEE International Conference on Robotics and Automation (ICRA)</em>, Stockholm, 2016, pp. 544-550.<br>
<br>
doi: 10.1109/ICRA.2016.7487177<br>
<br>
Abstract: This paper presents a novel approach for incremental semiparametric inverse dynamics learning. In particular, we consider the mixture of two approaches: Parametric modeling based on rigid body dynamics equations and nonparametric modeling based on incremental kernel methods, with no prior information on the mechanical properties of the system. The result is an incremental semiparametric approach, leveraging the advantages of both the parametric and nonparametric models. We validate the proposed technique learning the dynamics of one arm of the iCub humanoid robot.<br>
<br>
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7487177&isnumber=7487087<br>
</p>
</blockquote>
<p> </p>
<p><strong>Description</strong></p>
<p>The file "iCubDyn_2.0.mat" contains data collected from the right arm of the iCub humanoid robot, considering as input the positions, velocities and accelerations of the 3 shoulder joints and of the elbow joint, and as outputs the 3 force and 3 torque components measured by the six-axis F/T sensor in-built in the upper arm.</p>
<p>The dataset is collected at 10Hz at as the end-effector tracks circumferences with 10cm radius on the transverse (XY) and sagittal (XZ) planes (For more information on the iCub reference frames, see [4]) at approximately 0.6 m/s. The total number of points for each dataset is 10000, corresponding to approximately 17 minutes of continuous operation. Trajectories are generated by means of the Cartesian Controller presented in [5].</p>
<p> Input (X)</p>
<p> columns 1-4: Joint (3 shoulder joints + 1 elbow joint) positions<br>
columns 5-8: Joint (3 shoulder joints + 1 elbow joint) velocities<br>
columns 9-12: Joint (3 shoulder joints + 1 elbow joint) accelerations</p>
<p> Output (Y)</p>
<p> Columns 1-3: Measured forces (N) along the X, Y, Z axes by the force-torque (F/T) sensor placed in the upper arm<br>
Columns 4-6: Measured torques (N*m) along the X, Y, Z axes by the force-torque (F/T) sensor placed in the upper arm</p>
<p> </p>
<p><strong>Preprocessing</strong><br>
<br>
- Velocities and accelerations are computed by an Adaptive Window Polynomial Fitting Estimator, implemented through a least-squares based algorithm on a adpative window (see [2], [3]). Velocity estimation max window size: 16. Acceleration estimation max window size: 25.<br>
- Positions, velocities and accelerations are recorded at 9Hz and oversampled to 20 Hz via cubic spline interpolation.<br>
- Forces and torques are directly recorded at 20Hz.</p>
<p>This dataset was used in [1] for experimental purposes. See section IV therein for further details.</p>
<p>For more information, please contact:<br>
Raffaello Camoriano - [email protected]<br>
Silvio Traversaro - [email protected]</p>
<p> </p>
<p><strong>References</strong><br>
<br>
[1] Camoriano, Raffaello; Traversaro, Silvio; Rosasco, Lorenzo; Metta, Giorgio; Nori, Francesco, "Incremental Semiparametric Inverse Dynamics Learning", eprint arXiv:1601.04549, 01/2016<br>
[2] F. Janabi-Sharifi ; Dept. of Mech. Eng., Ryerson Polytech. Univ., Toronto, Ont., Canada ; V. Hayward ; C. -S. J. Chen, "Discrete-time adaptive windowing for velocity estimation", IEEE Transactions on Control Systems Technology, 1003 - 1009, Vol. 8, Issue 6, Nov 2000<br>
[3] https://github.com/robotology/icub-main/blob/master/src/libraries/ctrlLib/include/iCub/ctrl/adaptWinPolyEstimator.h<br>
[4] http://wiki.icub.org/wiki/ICubForwardKinematics<br>
[5] U. Pattacini; F. Nori; L. Natale; G. Metta; and G. Sandini; “An experimental evaluation of a novel minimum-jerk cartesian controller for humanoid robots,” in Intelligent Robots and Systems (IROS), 2010 IEEE/RSJ International Conference on, Oct 2010, pp. 1668–1674.</p>
Incremental semiparametric inverse dynamics learning
No full textThis paper presents a novel approach for incremental semiparametric inverse dynamics learning. In particular, we consider the mixture of two approaches: Parametric modeling based on rigid body dynamics equations and nonparametric modeling based on incremental kernel methods, with no prior information on the mechanical properties of the system. The result is an incremental semiparametric approach, leveraging the advantages of both the parametric and nonparametric models. We validate the proposed technique learning the dynamics of one arm of the iCub humanoid robot
Online semi-parametric learning for inverse dynamics modeling
No full textThis paper presents a semi-parametric algorithm for online learning of a robot inverse dynamics model. It combines the strength of the parametric and non-parametric modeling. The former exploits the rigid body dynamics equation, while the latter exploits a suitable kernel function. We provide an extensive comparison with other methods from the literature using real data from the iCub humanoid robot. In doing so we also compare two different techniques, namely cross validation and marginal likelihood optimization, for estimating the hyperparameters of the kernel function
Structured Prediction for CRiSP Inverse Kinematics Learning with Misspecified Robot Models
No full textWith the recent advances in machine learning, problems that traditionally would require accurate modeling to be solved analytically can now be successfully approached with data-driven strategies. Among these, computing the inverse kinematics of a redundant robot arm poses a significant challenge due to the non-linear structure of the robot, the hard joint constraints and the non-invertible kinematics map. Moreover, most learning algorithms consider a completely data-driven approach, while often useful information on the structure of the robot is available and should be positively exploited. In this work, we present a simple, yet effective, approach for learning the inverse kinematics. We introduce a structured prediction algorithm that combines a data-driven strategy with the model provided by a forward kinematics function - even when this function is misspecified - to accurately solve the problem. The proposed approach ensures that predicted joint configurations are well within the robot's constraints. We also provide statistical guarantees on the generalization properties of our estimator as well as an empirical evaluation of its performance on trajectory reconstruction tasks
La diagnosi funzionale mediante tecniche di gait analysis: esempio di quantificazione di parametri nel cammino della spina bifida
No full text- …
