Search Results for author: Dario Piga

Found 31 papers, 18 papers with code

Integrating Reinforcement Learning with Foundation Models for Autonomous Robotics: Methods and Perspectives

1 code implementation21 Oct 2024 Angelo Moroncelli, Vishal Soni, Asad Ali Shahid, Marco Maccarini, Marco Forgione, Dario Piga, Blerina Spahiu, Loris Roveda

We analyze the use of foundation models as action planners, the development of robotics-specific foundation models, and the mutual benefits of combining FMs with RL.

Reinforcement Learning (RL)

Enhanced Transformer architecture for in-context learning of dynamical systems

no code implementations4 Oct 2024 Matteo Rufolo, Dario Piga, Gabriele Maroni, Marco Forgione

Recently introduced by some of the authors, the in-context identification paradigm aims at estimating, offline and based on synthetic data, a meta-model that describes the behavior of a whole class of systems.

In-Context Learning Zero-Shot Learning

LightCPPgen: An Explainable Machine Learning Pipeline for Rational Design of Cell Penetrating Peptides

no code implementations31 May 2024 Gabriele Maroni, Filip Stojceski, Lorenzo Pallante, Marco A. Deriu, Dario Piga, Gianvito Grasso

The CPP predictive model works synergistically with an optimization algorithm, which is tuned to enhance computational efficiency while maintaining optimization performance.

Computational Efficiency

Model order reduction of deep structured state-space models: A system-theoretic approach

1 code implementation21 Mar 2024 Marco Forgione, Manas Mejari, Dario Piga

With a specific emphasis on control design objectives, achieving accurate system modeling with limited complexity is crucial in parametric system identification.

State Space Models

Synthetic data generation for system identification: leveraging knowledge transfer from similar systems

1 code implementation8 Mar 2024 Dario Piga, Matteo Rufolo, Gabriele Maroni, Manas Mejari, Marco Forgione

This paper addresses the challenge of overfitting in the learning of dynamical systems by introducing a novel approach for the generation of synthetic data, aimed at enhancing model generalization and robustness in scenarios characterized by data scarcity.

Synthetic Data Generation Transfer Learning

In-context learning of state estimators

no code implementations7 Dec 2023 Riccardo Busetto, Valentina Breschi, Marco Forgione, Dario Piga, Simone Formentin

State estimation has a pivotal role in several applications, including but not limited to advanced control design.

In-Context Learning

On the adaptation of in-context learners for system identification

1 code implementation7 Dec 2023 Dario Piga, Filippo Pura, Marco Forgione

In-context system identification aims at constructing meta-models to describe classes of systems, differently from traditional approaches that model single systems.

Meta-Learning

Gradient-based bilevel optimization for multi-penalty Ridge regression through matrix differential calculus

1 code implementation23 Nov 2023 Gabriele Maroni, Loris Cannelli, Dario Piga

Common regularization algorithms for linear regression, such as LASSO and Ridge regression, rely on a regularization hyperparameter that balances the tradeoff between minimizing the fitting error and the norm of the learned model coefficients.

Bilevel Optimization L2 Regularization +1

Shedding Light on the Ageing of Extra Virgin Olive Oil: Probing the Impact of Temperature with Fluorescence Spectroscopy and Machine Learning Techniques

no code implementations21 Sep 2023 Francesca Venturini, Silvan Fluri, Manas Mejari, Michael Baumgartner, Dario Piga, Umberto Michelucci

This work systematically investigates the oxidation of extra virgin olive oil (EVOO) under accelerated storage conditions with UV absorption and total fluorescence spectroscopy.

Split-Boost Neural Networks

no code implementations6 Sep 2023 Raffaele Giuseppe Cestari, Gabriele Maroni, Loris Cannelli, Dario Piga, Simone Formentin

The calibration and training of a neural network is a complex and time-consuming procedure that requires significant computational resources to achieve satisfactory results.

Data-Driven Computation of Robust Invariant Sets and Gain-Scheduled Controllers for Linear Parameter-Varying Systems

no code implementations4 Sep 2023 Manas Mejari, Ankit Gupta, Dario Piga

We present a direct data-driven approach to synthesize robust control invariant (RCI) sets and their associated gain-scheduled feedback control laws for linear parameter-varying (LPV) systems subjected to bounded disturbances.

Scheduling

From system models to class models: An in-context learning paradigm

6 code implementations25 Aug 2023 Marco Forgione, Filippo Pura, Dario Piga

Is it possible to understand the intricacies of a dynamical system not solely from its input/output pattern, but also by observing the behavior of other systems within the same class?

Decoder In-Context Learning

Neural State-Space Models: Empirical Evaluation of Uncertainty Quantification

1 code implementation13 Apr 2023 Marco Forgione, Dario Piga

Effective quantification of uncertainty is an essential and still missing step towards a greater adoption of deep-learning approaches in different applications, including mission-critical ones.

Deep Learning State Space Models +1

Learning Choice Functions with Gaussian Processes

1 code implementation1 Feb 2023 Alessio Benavoli, Dario Azzimonti, Dario Piga

We propose a Gaussian Process model to learn choice functions from choice-data.

Gaussian Processes

Direct identification of continuous-time linear switched state-space models

no code implementations4 Oct 2022 Manas Mejari, Dario Piga

The key idea for direct CT identification is based on an integral architecture consisting of an LSS model followed by an integral block.

State Space Models

Learning neural state-space models: do we need a state estimator?

5 code implementations26 Jun 2022 Marco Forgione, Manas Mejari, Dario Piga

In recent years, several algorithms for system identification with neural state-space models have been introduced.

State Space Models

On the adaptation of recurrent neural networks for system identification

1 code implementation21 Jan 2022 Marco Forgione, Aneri Muni, Dario Piga, Marco Gallieri

The system dynamics are then assumed to change, leading to an unacceptable degradation of the nominal model performance on the perturbed system.

Transfer Learning

Choice functions based multi-objective Bayesian optimisation

no code implementations15 Oct 2021 Alessio Benavoli, Dario Azzimonti, Dario Piga

In this work we introduce a new framework for multi-objective Bayesian optimisation where the multi-objective functions can only be accessed via choice judgements, such as ``I pick options A, B, C among this set of five options A, B, C, D, E''.

Bayesian Optimisation

A Model-Agnostic Algorithm for Bayes Error Determination in Binary Classification

no code implementations24 Jul 2021 Umberto Michelucci, Michela Sperti, Dario Piga, Francesca Venturini, Marco A. Deriu

This paper presents the intrinsic limit determination algorithm (ILD Algorithm), a novel technique to determine the best possible performance, measured in terms of the AUC (area under the ROC curve) and accuracy, that can be obtained from a specific dataset in a binary classification problem with categorical features {\sl regardless} of the model used.

Binary Classification

Deep learning with transfer functions: new applications in system identification

1 code implementation20 Apr 2021 Dario Piga, Marco Forgione, Manas Mejari

The dynamical operator is included as {the} last layer of a neural network in order to obtain the optimal one-step-ahead prediction error.

Deep Learning

A unified framework for closed-form nonparametric regression, classification, preference and mixed problems with Skew Gaussian Processes

1 code implementation12 Dec 2020 Alessio Benavoli, Dario Azzimonti, Dario Piga

In a recent contribution we showed that SkewGP and probit likelihood are conjugate, which allows us to compute the exact posterior for non-parametric binary classification and preference learning.

Active Learning Binary Classification +2

Computation of Parameter Dependent Robust Invariant Sets for LPV Models with Guaranteed Performance

no code implementations21 Sep 2020 Ankit Gupta, Manas Mejari, Paolo Falcone, Dario Piga

This paper presents an iterative algorithm to compute a Robust Control Invariant (RCI) set, along with an invariance-inducing control law, for Linear Parameter-Varying (LPV) systems.

Scheduling

Preferential Bayesian optimisation with Skew Gaussian Processes

no code implementations15 Aug 2020 Alessio Benavoli, Dario Azzimonti, Dario Piga

In this paper, we prove that the true posterior distribution of the preference function is a Skew Gaussian Process (SkewGP), with highly skewed pairwise marginals and, thus, show that Laplace's method usually provides a very poor approximation.

Bayesian Optimisation Gaussian Processes +2

Continuous-time system identification with neural networks: Model structures and fitting criteria

1 code implementation3 Jun 2020 Marco Forgione, Dario Piga

This paper presents tailor-made neural model structures and two custom fitting criteria for learning dynamical systems.

State Space Models

dynoNet: a neural network architecture for learning dynamical systems

3 code implementations3 Jun 2020 Marco Forgione, Dario Piga

This paper introduces a network architecture, called dynoNet, utilizing linear dynamical operators as elementary building blocks.

Skew Gaussian Processes for Classification

1 code implementation26 May 2020 Alessio Benavoli, Dario Azzimonti, Dario Piga

In spite of their success, GPs have limited use in some applications, for example, in some cases a symmetric distribution with respect to its mean is an unreasonable model.

Classification Gaussian Processes +1

Efficient Calibration of Embedded MPC

1 code implementation29 Nov 2019 Marco Forgione, Dario Piga, Alberto Bemporad

Model Predictive Control (MPC) is a powerful and flexible design tool of high-performance controllers for physical systems in the presence of input and output constraints.

Systems and Control Systems and Control Optimization and Control

Model structures and fitting criteria for system identification with neural networks

1 code implementation29 Nov 2019 Marco Forgione, Dario Piga

This paper focuses on the identification of dynamical systems with tailor-made model structures, where neural networks are used to approximate uncertain components and domain knowledge is retained, if available.

Active preference learning based on radial basis functions

no code implementations28 Sep 2019 Alberto Bemporad, Dario Piga

The algorithm described in this paper aims at reaching the global optimizer by iteratively proposing the decision maker a new comparison to make, based on actively learning a surrogate of the latent (unknown and perhaps unquantifiable) objective function from past sampled decision vectors and pairwise preferences.

Bayesian Optimization

Cannot find the paper you are looking for? You can Submit a new open access paper.