14 citations · 25 across the 16 of their papers we have counts for
16 papers
Probabilistic Verification of Recurrent Neural Networks for Single and Multi-Agent Reinforcement Learning
Luca Marzari, Enrico Marchesini
History-dependent policies induced by recurrent neural networks (RNNs) rely on latent hidden state dynamics, making verification in partially observable reinforcement learning (RL)…
On the Probabilistic Learnability of Compact Neural Network Preimage Bounds
Luca Marzari, Manuele Bicego, Ferdinando Cicalese +1
Although recent provable methods have been developed to compute preimage bounds for neural networks, their scalability is fundamentally limited by the #P-hardness of the problem. I…
Designing Control Barrier Function via Probabilistic Enumeration for Safe Reinforcement Learning Navigation
Luca Marzari, Francesco Trotti, Enrico Marchesini +1
Achieving safe autonomous navigation systems is critical for deploying robots in dynamic and uncertain real-world environments. In this paper, we propose a hierarchical control fra…
RobustX: Robust Counterfactual Explanations Made Easy
Junqi Jiang, Luca Marzari, Aaryan Purohit +1
The increasing use of Machine Learning (ML) models to aid decision-making in high-stakes industries demands explainability to facilitate trust. Counterfactual Explanations (CEs) ar…
Rigorous Probabilistic Guarantees for Robust Counterfactual Explanations
Luca Marzari, Francesco Leofante, Ferdinando Cicalese +1
We study the problem of assessing the robustness of counterfactual explanations for deep learning models. We focus on altering model parameters an…
ModelVerification.jl: a Comprehensive Toolbox for Formally Verifying Deep Neural Networks
Tianhao Wei, Hanjiang Hu, Luca Marzari +4
Deep Neural Networks (DNN) are crucial in approximating nonlinear functions across diverse applications, ranging from image classification to control. Verifying specific input-outp…
Improving Policy Optimization via -Retrain
Luca Marzari, Priya L. Donti, Changliu Liu +1
We present -retrain, an exploration strategy encouraging a behavioral preference while optimizing policies with monotonic improvement guarantees. To this end, we intro…
Scaling #DNN-Verification Tools with Efficient Bound Propagation and Parallel Computing
Luca Marzari, Gabriele Roncolato, Alessandro Farinelli
Deep Neural Networks (DNNs) are powerful tools that have shown extraordinary results in many scenarios, ranging from pattern recognition to complex robotic problems. However, their…
Enumerating Safe Regions in Deep Neural Networks with Provable Probabilistic Guarantees
Luca Marzari, Davide Corsi, Enrico Marchesini +2
Identifying safe areas is a key point to guarantee trust for systems that are based on Deep Neural Networks (DNNs). To this end, we introduce the AllDNN-Verification problem: given…
Constrained Reinforcement Learning and Formal Verification for Safe Colonoscopy Navigation
Davide Corsi, Luca Marzari, Ameya Pore +4
The field of robotic Flexible Endoscopes (FEs) has progressed significantly, offering a promising solution to reduce patient discomfort. However, the limited autonomy of most robot…
Safe Deep Reinforcement Learning by Verifying Task-Level Properties
Enrico Marchesini, Luca Marzari, Alessandro Farinelli +1
Cost functions are commonly employed in Safe Deep Reinforcement Learning (DRL). However, the cost is typically encoded as an indicator function due to the difficulty of quantifying…
Online Safety Property Collection and Refinement for Safe Deep Reinforcement Learning in Mapless Navigation
Luca Marzari, Enrico Marchesini, Alessandro Farinelli
Safety is essential for deploying Deep Reinforcement Learning (DRL) algorithms in real-world scenarios. Recently, verification approaches have been proposed to allow quantifying th…
The #DNN-Verification Problem: Counting Unsafe Inputs for Deep Neural Networks
Luca Marzari, Davide Corsi, Ferdinando Cicalese +1
Deep Neural Networks are increasingly adopted in critical tasks that require a high level of safety, e.g., autonomous driving. While state-of-the-art verifiers can be employed to c…
Verifying Learning-Based Robotic Navigation Systems
Guy Amir, Davide Corsi, Raz Yerushalmi +4
Deep reinforcement learning (DRL) has become a dominant deep-learning paradigm for tasks where complex policies are learned within reactive systems. Unfortunately, these policies a…
Curriculum Learning for Safe Mapless Navigation
Luca Marzari, Davide Corsi, Enrico Marchesini +1
This work investigates the effects of Curriculum Learning (CL)-based approaches on the agent's performance. In particular, we focus on the safety aspect of robotic mapless navigati…
Towards Hierarchical Task Decomposition using Deep Reinforcement Learning for Pick and Place Subtasks
Luca Marzari, Ameya Pore, Diego Dall'Alba +3
Deep Reinforcement Learning (DRL) is emerging as a promising approach to generate adaptive behaviors for robotic platforms. However, a major drawback of using DRL is the data-hungr…