04b Atto di convegno in volume

BANDITS WITH REPLENISHABLE KNAPSACKS: THE BEST OF BOTH WORLDS

The bandits with knapsacks (BwK) framework models online decision-making problems in which an agent makes a sequence of decisions subject to resource consumption constraints. The traditional model assumes that each action consumes a non-negative amount of resources and the process ends when the...
No-Regret Learning in Bilateral Trade via Global Budget Balance

Bilateral trade models the problem of intermediating between two rational agents - a seller and a buyer - both characterized by a private valuation for an item they want to trade. We study the online learning version of the problem, in which at each time step a new seller and buyer arrive and the...
Consistent Submodular Maximization

Maximizing monotone submodular functions under cardinality constraints is a classic optimization task with several applications in data mining and machine learning. In this paper we study this problem in a dynamic environment with consistency constraints: elements arrive in a streaming fashion and...
A Context-Aware Framework to Support Decision-Making in Production Planning

In the scope of Industry 4.0, this paper showcases the design and implementation of a context-aware decision-making framework that simulates the production planning of manufacturing activities by incorporating techniques from Process Mining, Business Process Simulation, and Visual Analytics. The...
An End-To-End Execution of a Logistic Process in an AI-Augmented Business Process Management System

This paper presents an end-to-end execution of a real-world business process (BP) in the logistics domain to illustrate how an AI-Augmented Business Process Management System (ABPMS) can increase BP automation compared to a traditional BPMS. In addition, we discuss concrete AI-based solutions for...
DIS-PIPE: A Tool for Data Pipeline Discovery

This paper presents DIS-PIPE, a software tool that leverages well-established process mining techniques to tackle the Data Pipeline Discovery (DPD) task. Data pipelines are composite steps that move data from disparate sources to some data consumers. While data travels through the pipeline, it can...
LTLf Goal-oriented Service Composition

Service compositions à la Roman model consist of realizing a virtual service by orchestrating suitably a set of already available services, where all services are described procedurally as (possibly nondeterministic) transition systems. In this paper, we study a goal-oriented variant of the service...
DanteLLM: Let’s Push Italian LLM Research Forward!

In recent years, the dominance of Large Language Models (LLMs) in the English language has become evident. However, there remains a pronounced gap in resources and evaluation tools tailored for non-English languages, underscoring a significant disparity in the global AI landscape. This paper seeks...
Rethinking Relevance: How Noise and Distractors Impact Retrieval-Augmented Generation

Retrieval-Augmented Generation (RAG) systems enhance the performance of Large Language Models (LLMs) by incorporating external information fetched from a retriever component. While traditional approaches prioritize retrieving “relevant” documents, our research reveals that these documents can be a...
Maintaining balance of mobile manipulators for safe pick-up tasks

This paper presents a novel method to maintain the dynamic balance of a Mobile Manipulator (MM) during the pick-up of heavy objects. The approach entails the generation of a preliminary reach-to-grasp trajectory, which is subsequently refined by an Optimization-Based Controller (OBC) formulated as...