This document includes links to files in ps and pdf format.
Problems in downloading the files? Please, send me an email: campi@ing.unibs.it


Conference paper


S D. Paccagnan, M.C. Campi, S. Garatti

The Pick-to-Learn Algorithm: Empowering Compression for Tight Generalization Bounds and Improved Post-training Performance.

37th Conference on Neural Information Processing Systems - Neurips2022


S Garatti, M.C. Campi.

Compression at the service of learning: a case study for the Guaranteed Error Machine.

61st IEEE Conference on Decision and Control (CDC), 2135-2140, 2022


S Garatti, M.C. Campi.

On the consistency of the risk evaluation in the scenario approach.

60th IEEE Conference on Decision and Control (CDC), 1468-1473, 2021


S Garatti, MC Campi.
The risk of making decisions from data through the lens of the scenario approach.
19th IFAC Symposium on System Identificaiton SYSID, 607-612, 2021.

 

MC Campi, S Garatti

Scenario optimization with relaxation: a new tool for design and application to machine learning problems.

59th IEEE Conference on Decision and Control (CDC), 2463-2468, 2020.

 

A.Carč, M.C.Campi, F.A.Ramponi, S.Garatti, A.T.J.R.Cobbenhagen

A study of majority-voting classifiers with guarantee on the probability of error.

21st IFAC World Congress, Berlin, 1013-1018, 2020

 

S Garatti, MC Campi.
Learning for control: a Bayesain scenario approach.

IEEE 58th Conference on Decision and Control (CDC), 1772-1777, 2019.

D Paccagnan, MC Campi.
The scenario approach meets uncertain game theory and variational inequalities.

IEEE 58th Conference on Decision and Control (CDC), 6124-6129, 2019.

S Garatti, MC Campi.
Complexity-based modulation of the data-set in scenario optimization.

18th European Control Conference (ECC), 1386-1391, 2019.

A. Carč, F. Torricelli, M.C. Campi, S. Savaresi.
A Toolbox for Virtual Reference Feedback Tuning (VRFT).
In Proc. of the European Control Conference 2019, Napoli, Italy, 4252-4257, 2019.

A. Carč, G. Pillonetto, M.C. Campi.
Uncertainty bounds for kernel-based regression: a Bayesian SPS approach.
IEEE International Workshop on Machine Learning for Signal Processing, MLSP, Aalborg, Denmark, pp.17–20, 2018.

F. Baronio, M. Baronio, M.C. Campi, A. Carč, S. Garatti, G. Perone.
Ventricular defibrillation: Classification with G.E.M. and a roadmap for future investigations.
In Proc. 56th Conf. on Decision and Control, Melbourne, Australia, pages 2718-2723, 2018.

S. Formentin, S. Garatti, M.C. Campi, S. Savaresi.
Tuning regularization via scenario optimization.
In Proc. 56th Conf. on Decision and Control, Melbourne, Australia, pages 635-640, 2018.

G. Pillonetto, A. Carč, M.C. Campi.
Kernel-based SPS.

IFAC-PapersOnline, Volume 51, Issue 15, 1 January 2018, Pages 31-36.
(18th IFAC Symposium on System Identification - SYSID, 2018).

A. Caré, M.C. Campi, Balázs Csáji, E. Weyer.
Undermodelling detection with sign-perturbed sums.
In Proc. 20th World IFAC Congress, Toulouse, France, 2799-2804, 2017.

A. Caré, B.C. Csaji, M.C. Campi.
Sign-perturbed sums (SPS) with asymmetric noise: robustness analysis and robustification techniques.
In Proc. 55th Conf. on Decision and Control, Las Vegas, USA, 2016.

S Trimpe, M.C. Campi.
On the choice of the event trigger in event-based estimation.
International Conference on Event-Based Control, Communication, and Signal Processing (EBCCSP) pp: 1-8, 2015.


M.C. Campi, S. Garatti, F. Ramponi
Non-convex scenario optimization with application to system identification.

Proceedings of the 54th IEEE Conference on Decision and Control, Osaka, Japan, 2015.

V. Volpe, B.C. Csaji, A. Caré and M.C. Campi.
Sign-Perturbed Sums (SPS) with instrumental variables for the identification of ARX systems.
In Proc. 54th Conf. on Decision and Control, Osaka, Japan, 2015.

S. Formentin, M.C. Campi, S. Savaresi.
Virtual Reference Feedback Tuning for Industrial PID Controllers.
In Proc. 19th World IFAC Congress, Cape Town, South Africa, 2014.

P. Panciatici, M.C. Campi, S. Garatti, S.H. Low, D.K. Molzahn, A.X. Sun, L. Wehenkel
Advanced optimization methods for power systems.
In Proc. 18th Power Systems Computation Conference, Wroclaw, Poland, 2014.

A. Caré, S. Garatti, M.C. Campi
Empirical cost distribution: a scenario approach to the construction of probability boxes with application to channel equalization.
In Proc. 13th European Control Conference, Strasbourg, France, 2014.

E. Weyer, B.C. Csaji and M.C. Campi.
Guaranteed Non-Asymptotic Confidence Ellipsoids for FIR Systems.
In Proc. 52nd Conf. on Decision and Control, Firenze, Italy, 2013.

A. Care', S. Garatti and M.C. Campi.
Least Squares Estimates and the Coverage of Least Squares Costs.
In Proc. 52nd Conf. on Decision and Control, Firenze, Italy, 2013.

B.C. Csaji, M.C. Campi and E. Weyer.
Sign-Perturbed Sums (SPS): A Method for Constructing Exact Finite-Sample Confidence Regions for General Linear Systems.
In Proc. 51st Conf. on Decision and Control, Maui, Hawaii, USA , 2012.

M.C. Campi, B.C. Csaji, S. Garatti and E. Weyer.
Certified system  identification: towards distribution-free results.
Plenary address. In Proc. Symposium on System Identification, SYSID 2012, Brussels, Belgium, 245-255, 2012.

A. Care', S. Garatti and M.C. Campi.
FAST: an algorithm for the scenario approach with reduced sample complexity.
In Proc. 18th World IFAC Congress, Milano, Italia, 2011.

A. Care', S. Garatti and M.C. Campi.
Randomized min-max optimization: the exact risk of multiple cost levels.
In Proc. 50th Conf. on Decision and Control, Orlando, FL, USA, 7794-7799, 2011.

E. Weyer and M.C. Campi.
Prediction, filtering and smoothing using LSCR: state estimation algorithms with guaranteed confidence sets.
In Proc. 50th Conf. on Decision and Control, Orlando, FL, USA, 2589-2594, 2011.

E. Weyer and M.C. Campi.
State estimation algorithms with guaranteed confidence intervals for first order systems.
In Proc. 9th IEEE International Conference on Control and Automation (ICCA), Santiago, Chile, 2011.

M.C. Campi and S. Garatti.
Variable Robustness Control: Principles and Algorithms.
Semi-plenary address. In Proc. 19th Symposium on Mathematical Theory of Networs and Sytems, Budapest, Hungary, 2010.

E. Weyer, S. Ko and M.C. Campi.
Finite sample properties of system identification with quantized output data.
In Proc. 48th Conf. on Decision and Control, Shanghai, China, 2009.

S. Bittanti, M.C. Campi and M. Prandini.
Adaptation and the effort needed to adapt.
In Proc. 48th Conf. on Decision and Control, Shanghai, China, 2009.

E. Weyer, S. Ko and M.C. Campi. 
A randomised subsampling method for change detection.

In Proc. SafeProcess'09, Barcelona, Spain, 2009.
(Paul Frank award finalist)

K.J. Layton, E. Weyer, and M.C. Campi. 
Online algorithms for the construction of guaranteed confidence sets for the parameters of time-varying systems.
In Proc. Symposium on System Identification, SYSID 2009, Saint Malo, pp. 426-431, 2009.

S. Garatti and M.C. Campi. 
L-infinity layers and the probability of false prediction.

In Proc. Symposium on System Identification, SYSID 2009, Saint Malo, 2009.

M.C. Campi, S. Garatti and M. Prandini.
The scenario approach for systems and control design.
In Proc. 17th World IFAC Congress, Seoul, Korea, 381-389, 2008.
(also appeared in: Plenary Papers, Milestone Reports & Selected Survey Papers (M.J. Chung and P. Misra eds.). IFAC WC 2008,  pages 180-188, 2008.)

M.C. Campi
How much do we need to experiment in order to adapt?
In Proc. of the European Control Conference 2007, Kos, Greece, 5556-5557, 2007.

M.C. Campi, Sangho Ko and Erik Weyer
Non-asymptotic confidence regions for model parameters in the presence of unmodelled dynamics.
In Proc. 46th Conf. on Decision and Control, New Orleans, LA, USA, 2007.

M.C. Campi and Giuseppe Calafiore.
New results on the scenario design approach.
In Proc. 46th Conf. on Decision and Control, New Orleans, LA, USA, 2007.

M.C. Campi and Simone Garatti
Modulating robustness in robust control: making it easy through randomization.
In Proc. 46th Conf. on Decision and Control, New Orleans, LA, USA, 2007.

Maria Prandini and M.C. Campi. 
A new approach to controller design in presence of constraints.

In Proc. 46th Conf. on Decision and Control, New Orleans, LA, USA, 2007.


Sangho Ko, Erik Weyer and M.C. Campi. 
Non-asymptotic uncertainty assessment of frequency responses using the LSCR approach.

In Proc. 2007 Conf. on Control, Automation and Systems, ICCAS 2007, Seoul, Korea, 2007.


M.C. Campi, Toshiharu Sugie and Fumitoshi Sakai. 
Iterative identification method for linear continuous-time systems.
In Proc. 45th Conf. on Decision and Control, San Diego, CA, USA, 817-822, 2006.

M.C. Campi and E. Weyer.
Non-asymptotic confidence sets for input-output transfer functions.
In Proc. 45th Conf. on Decision and Control, San Diego, CA, USA, 157-162, 2006.

M.C. Campi and E. Weyer.
Identification with finitely many data points: the LSCR approach
Semi-plenary address. In Proc. Symposium on System Identification, SYSID 2006, New Castle, Australia, 2006.

M.C. Campi and S. Savaresi.
Virtual reference feedback tuning for nonlinear systems.
In Proc. 44th Conf. on Decision and Control, Seville, Spain, 6608-6613, 2005.

M. Dalai, E. Weyer and M.C. Campi.
Parametric identification of nonlinear systems: Guaranteed confidence regions.
In Proc. 44th Conf. on Decision and Control, Seville, Spain, 6418-6423, 2005.

G. Calafiore and M.C. Campi.
On two-stage portfolio allocation problems with affine recourse.
In Proc. 44th Conf. on Decision and Control, Seville, Spain, 8042-8046, 2005.

S. Garatti, M.C. Campi and S. Bittanti.
Introducing robustness in iterative control.
In Proc. 44th Conf. on Decision and Control, Seville, Spain, 3111-3116, 2005.

T. Katayama, T. McKelvey, A. Sano, C. Cassandras and M.C. Campi.
Trends in systems and signals.
Status report prepared by the IFAC coordinating committee on systems and signals.

In Proc. 16th World IFAC Congress, Prague, 2005.

E. Weyer and M.C. Campi.
Global non-asymptotic confidence sets for general linear models.
In Proc. 16th World IFAC Congress, Prague, 2005.

M.C. Campi, G. Calafiore and S. Garatti.
New results on the identification of interval predictor models.

In Proc. 16th World IFAC Congress, Prague, 2005.

M.C. Campi and E. Weyer.
Non-asymptotic confidence sets for general linear system identification models. 

In Proc. 1st workshop on Advanced Control Circuits and Systems, Cairo, 2005. 

G. Calafiore and M.C. Campi.
A new bound on the generalization rate of sampled convex programs.
In Proc. 43rd Conf. on Decision and Control, Atlantis, Bahamas, USA , 2004.

S. Bittanti, M.C. Campi and S. Garatti.
Some critical implementation issues in iterative robust control design.
In Proc. IFAC Workshop on Adaptation and Learning in Control and Signal Processing (ALCOSP 2004), Yokohama, Japan, 2004.

M.C. Campi and E. Weyer.
Non-asymptotic confidence sets for the parameters of ARMAX models.
In Proc. IFAC Workshop on Adaptation and Learning in Control and Signal Processing (ALCOSP 2004), Yokohama, Japan, 2004.

M.C. Campi and E. Weyer
Estimation of confidence regions for the parameters of ARMA models - guaranted non-asymptotic results.
In Proc. 42st Conf. on Decision and Control, Maui, Hawaii, USA , 2003.

S. Garatti, S. Bittanti and M.C. Campi
Model quality assessment for Instrumental Variable methods: use of the asymptotic theory in practice.
In Proc. 42st Conf. on Decision and Control, Maui, Hawaii, USA , 2003.

G. Calafiore and M.C. Campi
Robust Convex Programs: Randomized Solutions and Applications in Control.
In Proc. 42st Conf. on Decision and Control, Maui, Hawaii, USA , 2003.

Su Ki Ooi, E. Weyer and M.C. Campi.
Finite sample quality assessment of system identification models of irrigation channels.
2003 IEEE Conference on Control Applications, Istanbul, 2003.

M.C. Campi, Su Ki Ooi and E. Weyer.
Non-asymptotic quality assessment of generalised FIR models.
In Proc. 41st Conf. on Decision and Control, Las Vegas, USA, 2002.

G. Calafiore and M.C. Campi.
Interval predictors for unknown dynamical systems: an assessment of reliability.
In Proc. 41st Conf. on Decision and Control, Las Vegas, USA, 2002.

M.C. Campi, Joao P. Hespanha, Maria Prandini.
Cautious hierarchical switching control of stochastic linear systems.
Int. Symposium on the Mathematical Theory of Network and Systems, Notre-Dame, Indiana, 2002.

S.Bittanti, M.C. Campi and S. Garatti.
An iterative controller design scheme based on average robust control.
In Proc. 15th World IFAC Congress, Barcelona, 2002.

G. Calafiore, M.C. Campi and L. El Ghaui.
Identification of reliable predictor models for unknown systems: a data-consistency aproach based on learning theory.
In Proc. 15th World IFAC Congress, Barcelona, 2002.

M.C. Campi, E. Weyer and Su Ki Ooi.
Nonasymptotic quality assessment of identified models.
In Proc. 15th World IFAC Congress, Barcelona, 2002.

A. Lecchini, M.C. Campi and S.M Savaresi.
Sensitivity shaping via virtual reference feedback tuning.
In Proc. 40th Conf. on Decision and Control, Orlando, pages 750-755, 2001.

M. Prandini and M.C. Campi.
Logic-based switching for the stabilization of stochastic systems in presence of unmodeled dynamics.
In Proc. 40th Conf. on Decision and Control, Orlando, pages 393-398, 2001.

A. Lecchini, M.C. Campi and S.M Savaresi.
Virtual reference feedback tuning for two degree of freedom controllers.
European Control Conferece ECC 2001, Porto, 2001.

S.Bittanti, M.C. Campi and M. Prandini.
Penalized identification for self-tuning control: an overview.
IFAC ALCOSP Workshop, Como, 2001.

A. Lecchini, M.C. Campi and S.M Savaresi.
Virtual reference feedback tuning: a new framework for data-based design of PID  and  linear  controllers.
IFAC ALCOSP Workshop, Como, 2001.

M.C. Campi and M. Vidyasagar.
Learning with prior information.
In the Proceedings of the 2000 IEEE International Symposium on Circuits and Systems (ISCAS), Geneva, pages 255-258, vol.3, 2000.

M.C. Campi, R. Leonardi and L.A. Rossi.
Pole placement techniques for a class of IIR blind equalizers.
In Proc. EUSIPCO, Tampere, 2000.

M.C. Campi, A. Lecchini and S.M Savaresi.
Virtual reference feedback tuning (VRFT): a new direct approach to the design of feedback controllers.
In Proc. 39th Conf. on Decision and Control, Sydney, pages 623-628, 2000.

E. Weyer and M.C. Campi.
Non-asymptotic confidence ellipsoids for the least squares estimate.
In Proc. 39th Conf. on Decision and Control, Sydney, pages 2688-2693, 2000.

S. Bittanti, M.C. Campi and L. Guo.
Persistence of excitation properties for the identification of time-varying systems.
In Proc. 39th Conf. on Decision and Control, Sydney, pages 680-685, 2000.

M. Prandini e M.C. Campi.
A self-optimizing adaptive LQG control scheme for input-output systems.
In Proc. 39th Conf. on Decision and Control, Sydney, pages 1110-1115, 2000.

E. Weyer and M.C. Campi.
Finite sample properties in system identification methods.
In Proc. 38th Conf. on Decision and Control, Phoenix, 1999.

S. Bittanti and M.C. Campi.
Penalized identification methods for self-tuning control.
In Symposium on Adaptive Systems (SPAS), St.Petersburg, pages 17-22, 1999.

M.C. Campi, R. Leonardi and L. Rossi.
Generalized super-exponential method for blind equalization using Kautz filters.
In IEEE Signal Processing Workhop on Higher Order Statistics, Caesarea, pages 107-111, 1999.

M.C. Campi and M. Prandini.
Randomized algorithms for the synthesis of adaptive controllers.
Int. Symposium on the Mathematical Theory of Network and Systems, Padova, pages 723-726, 1998.

M.C. Campi, A. Lecchini, M. Pardo and S. Savaresi.
Iterative feedback tuning: a direct approach.
In Int. Symposium on the Mathematical Theory of Network and Systems, Padova, pages 719-722, 1998.

M. Prandini, S. Bittanti and M.C. Campi.
The stabilizability issue in least squares estimation: a penalized approach.
In Int. Symposium on the Mathematical Theory of Network and Systems, Padova, pages 855-858, 1998.

S. Bittanti, M.C. Campi and S. Savaresi.
Notch filters versus funnel filters in frequency estimation.
In Proc. Int. Symposium on Adaptive Control and Signal Processing, Glasgow, 1998.

S. Bittanti and M.C. Campi.
Least squares identification of time varying parameters: issues and trends.
In Proc. 11th IFAC Symposium on System Identification, Fukuoka, 1997.

M.C. Campi.
Achieving optimality in adaptive control: the Bet On the Best approach.
In Proc. 36th Conf. on Decision and Control, San Diego, 1997.

M. Prandini and M.C. Campi.
Adaptive pole placement by means of a simple, singularity free, identification algorithm.
In Proc. 36th Conf. on Decision and Control, San Diego, 1997.

M.C. Campi and P.R. Kumar.
Optimal adaptive control of an LQG system.
In Proc. 35th Conf. on Decision and Control, Kobe, pages 349-353, 1996.

M.C. Campi and P.R. Kumar.
Learning dynamical systems in a stationary environment.
In Proc. 35th Conf. on Decision and Control, Kobe, pages 2308-2311, 1996.

S. Bittanti and M.C. Campi.
Do self-tuning control systems tune?
In Proc. 13th World IFAC Congress, San Francisco, pages 115-120, 1996.

S. Bittanti, M.C. Campi, and M. Prandini.
Self-tuning control based on penalized least squares identification techniques.
In Proc. 35th Conference on Decision and Control, Kobe, pages 62-63, 1996.

M.C. Campi and M. Prandini.
Singularity free adaptive LQG control: stability and optimality.
In Int. Symposium on the Mathematical Theory of Network and Systems, St. Louis, 1996.

M. Prandini, M.C. Campi, and R. Leonardi.
A new algorithm for the automatic search of the best delay in blind equalization.
In IEEE Digital Signal Processing Workshop, Loen, pages 458-461, 1996.

M. Prandini, M.C. Campi, and R. Leonardi.
New results on the evaluation of equalizers performance.
In IEEE Digital Signal Processing Workshop, Loen, pages 462-465, 1996.

P. Belli, S. Bittanti, P. Bolzern, M.C. Campi, A. De Marco, A. Ferretti, S. Malloggi, and W. Prandoni.
A control system for nitrogen oxides pollution abatement by SCR (selective catalytic reduction).
In 13th World IFAC Congress, San Francisco, pages 13-18, 1996.

S. Bittanti, P. Bolzern, M.C. Campi, A. De Marco, R. Panseri, G. Poncia, and W. Prandoni.
A model of a bubbling fluidized bed combustor for the estimation of the char mass via extended kalman filter.
In 13th World IFAC Congress, San Francisco, pages 67-72, 1996.

F. Lorito and M.C. Campi.
Generalized forgetting techniques in adptive RLS identification.
In CESA '96 IMAX Multiconference, Lille, 1996.

S. Bittanti and M.C. Campi.
Self-optimality of adaptive control systems based on the certainty equivalence principle.
In Proc. 5th IFAC Symposium on Adaptive Systems in Control and Signal Processing , Budapest, 1995.

S. Bittanti, M.C. Campi, and S. Savaresi.
Unbiased estimation of a sinusoid in noise via notch filters.
In 5th IFAC Symposium on Adaptive Systems in Control and Signal Processing, Budapest, pages 165-170, 1995.

R. Bardelli, S. Bittanti, P. Bolzern, M.C. Campi, E. Carugati, A. De Marco, L. Piroddi, and W. Prandoni.
Application of the extended Kalman filter to the estimation of the char mass in a fluidized bed combustor.
In 10th IFAC Symposium on System Identification, Copenhagen, pages 247-252, 1994.

S. Bittanti and M.C. Campi.
Least squares identification of autoregressive models with time varying  parameters.
In Proc. 33rd Conference on Decision and Control, Lake Buena Vista, pages 3610-3611, 1994.

S. Bittanti and M.C. Campi.
Adaptive RLS identification of time-varying parameters - when is the error bounded?
In 10th IFAC Symposium on System Identification, Copenhagen, pages 367-372, 1994.

M.C. Campi and M. James.
Risk-sensitive control: a bridge between $h_\infty$ and $h_2$ control.
In Proc. 32nd Conf. on Decision and Control, San Antonio, pages 2179-2180, 1993.

M.C. Campi and F. Lorito.
Adaptive RLS identification with generalized forgetting factor.
In Int. Symposium on the Mathematical Theory of Network and Systems, Regensburg, pages 91-92, 1993.

M.C. Campi, W.S. Lee, and B.D.O. Anderson.
Filters for internal model control design for unstable plants.
In Proc. 32nd Conf. on Decision and Control, San Antonio, pages 1343-1348, 1993.

S. Bittanti and M.C. Campi.
A fake information matrix approach to the analysis of finite memory RLS identification techniques.
In Proc. 32nd Conference on Decision and Control, San Antonio, pages 2026-2031, 1993.

S. Bittanti, M.C. Campi, and F.Lorito.
ICSS: A program for the identification and control of time varying systems.
In European Simulation Multiconference, pages 588-592, 1992.

S. Bittanti and M.C. Campi.
Tuning the forgetting factor in RLS identification algorithms.
In Proc. 30th Conference on Decision and Control, Brighton, pages 1688-1689, 1991.

S.Bittanti and M.C. Campi.
Mean square convergence of an adaptive RLS algorithm with stochastic excitation.
In Proc. 28th Conference on Decision and Control, Tampa, pages 1946-1951, 1989.

S. Bittanti, P. Bolzern, M.C. Campi, and E. Coletti.
Deterministic convergence analysis of RLS estimators with different forgetting factors.
In Proc. 27th Conference on Decision and Control, Austin, pages 1530-1531, 1988.