Search Results for author: Alessio Russo

Found 15 papers, 6 papers with code

Constrained Deep Reinforcement Learning for Fronthaul Compression Optimization

no code implementations26 Sep 2023 Axel Grönland, Alessio Russo, Yassir Jedra, Bleron Klaiqi, Xavier Gelabert

In the Centralized-Radio Access Network (C-RAN) architecture, functions can be placed in the central or distributed locations.

Quantization reinforcement-learning

What influences occupants' behavior in residential buildings: An experimental study on window operation in the KTH Live-In Lab

no code implementations16 Jul 2023 Mahsa Farjadnia, Angela Fontan, Alessio Russo, Karl Henrik Johansson, Marco Molinari

Window-opening and window-closing behaviors play an important role in indoor environmental conditions and therefore have an impact on building energy efficiency.

On the Sample Complexity of Representation Learning in Multi-task Bandits with Global and Local structure

1 code implementation28 Nov 2022 Alessio Russo, Alexandre Proutiere

Arms consist of two components: one that is shared across tasks (that we call representation) and one that is task-specific (that we call predictor).

Representation Learning

Analysis and Detectability of Offline Data Poisoning Attacks on Linear Dynamical Systems

3 code implementations16 Nov 2022 Alessio Russo

In recent years, there has been a growing interest in the effects of data poisoning attacks on data-driven control methods.

Data Poisoning

Model Based Residual Policy Learning with Applications to Antenna Control

no code implementations16 Nov 2022 Viktor Eriksson Möllerstedt, Alessio Russo, Maxime Bouton

Non-differentiable controllers and rule-based policies are widely used for controlling real systems such as telecommunication networks and robots.

Reinforcement Learning (RL)

Self-Tuning Tube-based Model Predictive Control

no code implementations2 Oct 2022 Damianos Tranos, Alessio Russo, Alexandre Proutiere

We present Self-Tuning Tube-based Model Predictive Control (STT-MPC), an adaptive robust control algorithm for uncertain linear systems with additive disturbances based on the least-squares estimator and polytopic tubes.

Model Predictive Control

Tube-Based Zonotopic Data-Driven Predictive Control

1 code implementation7 Sep 2022 Alessio Russo, Alexandre Proutiere

We present a novel tube-based data-driven predictive control method for linear systems affected by a bounded addictive disturbance.

Computational Efficiency

Balancing detectability and performance of attacks on the control channel of Markov Decision Processes

1 code implementation15 Sep 2021 Alessio Russo, Alexandre Proutiere

In such an attack, drawing inspiration from adversarial examples used in supervised learning, the amplitude of the adversarial perturbation is limited according to some norm, with the hope that this constraint will make the attack imperceptible.

Reinforcement Learning (RL)

Some Ethical Issues in the Review Process of Machine Learning Conferences

no code implementations1 Jun 2021 Alessio Russo

Recent successes in the Machine Learning community have led to a steep increase in the number of papers submitted to conferences.

BIG-bench Machine Learning

Data-Driven Control and Data-Poisoning attacks in Buildings: the KTH Live-In Lab case study

no code implementations10 Mar 2021 Alessio Russo, Marco Molinari, Alexandre Proutiere

This work investigates the feasibility of using input-output data-driven control techniques for building control and their susceptibility to data-poisoning techniques.

Data Poisoning

Poisoning Attacks against Data-Driven Control Methods

no code implementations10 Mar 2021 Alessio Russo, Alexandre Proutiere

This paper investigates poisoning attacks against data-driven control methods.


Minimizing Information Leakage of Abrupt Changes in Stochastic Systems

1 code implementation2 Mar 2021 Alessio Russo, Alexandre Proutiere

In contrast to previous work on privacy, we study the problem for an online sequence of data.

Optimal Attacks on Reinforcement Learning Policies

no code implementations31 Jul 2019 Alessio Russo, Alexandre Proutiere

Finally, we show that from the main agent perspective, the system uncertainties and the attacker can be modeled as a Partially Observable Markov Decision Process.

reinforcement-learning Reinforcement Learning (RL)

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