Pier Luca Lanzi

• PhD
• Professor (Full) at Politecnico di Milano

We present an approach to identify and track the evolution of niches in XCS that can be applied to any XCS model and any problem. It exploits the underlying principles of the evolutionary component of XCS, and therefore, it is independent of the representation used. It also employs information already available in XCS and thus requires minimal modi...

Physics Playground is an immersive Virtual Reality (VR) application designed for educational purposes, featuring a virtual laboratory where users interact with various physics phenomena through guided experiments. This study aims to evaluate the application's design and educational content to facilitate its integration into classroom settings. A qu...

Among all artificial intelligence (AI) techniques and methods, machine learning (ML) includes all those approaches that allow computers to learn from data without being explicitly programmed. ML has been extensively applied to medical imaging, and the last few years have seen the rise of deep learning (DL), which is an area of ML focused on the app...

Large language models (LLMs) have taken the scientific world by storm, changing the landscape of natural language processing and human-computer interaction. These powerful tools can answer complex questions and, surprisingly, perform challenging creative tasks (e.g., generate code and applications to solve problems, write stories, pieces of music,...

To identify the best transfer learning approach for the identification of the most frequent abnormalities on chest radiographs (CXRs), we used embeddings extracted from pretrained convolutional neural networks (CNNs). An explainable AI (XAI) model was applied to interpret black-box model predictions and assess its performance. Seven CNNs were train...

We present an educational application of virtual reality that we created to help students gain an in-depth understanding of the internal structure of crystals and related key concepts. Teachers can use it to give lectures to small groups (10-15) of students in a shared virtual environment, both remotely (with teacher and students in different locat...

We present the evaluation of the V-Arcade serious games framework to support hand, wrist and forearm rehabilitation for children with Juvenile Idiopathic Arthritis (JIA). We evaluated the framework in collaboration with the therapists of the De Marchi Pediatric Clinic in Milan. We provided the software at home to three children with JIA, each child...

Minimalist game design was introduced a decade ago as a general design principle with a list of key properties for minimalist games: basic controls, simple but aesthetically pleasing visuals, interesting player choices with vast possibility spaces, and sounds that resonate with the design. In this paper, we present an experiment we did to explore m...

Purpose To identify the best transfer learning approach in the identification of the most frequent abnormalities on chest radiograph (CXR) using embeddings extracted from pre-trained convolutional neural networks (CNNs). Explainable AI (XAI) model was applied to interpret black-box models’ predictions and assess its performance. Methods and Materi...

We apply the extension of Monte Carlo Tree Search for single player games (SP-MCTS) to Sokoban and compare its performance to a solver integrating Iterative Deepening A* (IDA*) with several problem-specific heuristics. We introduce two extensions of MCTS to deal with some of the challenges that Sokoban poses to MCTS methods, namely, the reduced sea...

Purpose The present scoping review aims to assess the non-inferiority of distributed learning over centrally and locally trained machine learning (ML) models in medical applications. Methods We performed a literature search using the term “distributed learning” OR “federated learning” in the PubMed/MEDLINE and EMBASE databases. No start date limit...

Deep Learning has established in the latest years as a successful approach to address a great variety of tasks. Healthcare is one of the most promising field of application for Deep Learning approaches since it would allow to help clinicians to analyze patient data and perform diagnoses. However, despite the vast amount of data collected every year...

In recent years new technologies have been developed to track user's motions, as the Leap Motion Controller for hands and wrist tracking. We propose a framework for children's hand, wrist and forearm rehabilitation at home and at hospital designed and developed in collaboration with the therapists of the De Marchi Pediatric Clinic in Milan. The fra...

Chest X-ray (CXR) is perhaps the most frequently-performed radiological investigation globally. In this work, we present and study several machine learning approaches to develop automated CXR diagnostic models. In particular, we trained several Convolutional Neural Networks (CNN) on the CheXpert dataset, a large collection of more than 200k CXR lab...

Patients undergoing physical rehabilitation therapy must perform series of exercises regularly over a long period of time to improve, or at least not to worsen, their condition. Rehabilitation can easily become boring because of the tedious repetition of simple exercises, which can also cause mild pain and discomfort. As a consequence, patients oft...

Space debris represents a threat to space missions and operational satellites. Failing to control its growth might lead to the inability to use near-Earth space. However, this issue is still largely unknown to most people. In this paper, we present an educational experience in virtual reality created to raise awareness about the problem of space de...

We present the design of a competitive artificial intelligence for Scopone, a popular Italian card game. We compare rule-based players using the most established strategies (one for beginners and two for advanced players) against players using Monte Carlo Tree Search (MCTS) and Information Set Monte Carlo Tree Search (ISMCTS) with different reward...

We present the design of a competitive artificial intelligence for Scopone, a popular Italian card game. We compare rule-based players using the most established strategies (one for beginners and two for advanced players) against players using Monte Carlo Tree Search (MCTS) and Information Set Monte Carlo Tree Search (ISMCTS) with different reward...

Rehabilitation is a painful and tiring process involving series of exercises that patients must repeat over a long period. Unfortunately, patients often grow bored, frustrated, and lose motivation making rehabilitation less effective. In the recent years video games have been widely used to implement rehabilitation protocols so as to make the proce...

We present the design of a competitive artificial intelligence for {Scopone}, a popular Italian card game. We compare rule-based players using the most established strategies (one for beginners and two for advanced players) against players using Monte Carlo Tree Search (MCTS) and Information Set Monte Carlo Tree Search (ISMCTS) with different rewar...

Rehabilitation is a painful and tiring process involving series of exercises that patients must repeat over a long period. Unfortunately, patients often grow bored, frustrated, and lose motivation making rehabilitation less effective. In the recent years video games have been widely used to implement rehabilitation protocols so as to make the proce...

We applied Generative Adversarial Networks (GANs) to learn a model of DOOM levels from human-designed content. Initially, we analysed the levels and extracted several topological features. Then, for each level, we extracted a set of images identifying the occupied area, the height map, the walls, and the position of game objects. We trained two GAN...

The design of video game levels is a complex and critical task. Levels need to elicit fun and challenge while avoiding frustration at all costs. In this paper, we present a framework to assist designers in the creation of levels for 2D platformers. Our framework provides designers with a toolbox (i) to create 2D platformer levels, (ii) to estimate...

We applied Generative Adversarial Networks (GANs) to learn a model of DOOM levels from human-designed content. Initially, we analysed the levels and extracted several topological features. Then, for each level, we extracted a set of images identifying the occupied area, the height map, the walls, and the position of game objects. We trained two GAN...

Developmental dyslexia is a specific learning disorder of neurobiological origin that causes a reading impairment. Since music and language share common mechanisms and the core deficit underlying dyslexia has been identified in difficulties in dynamic and rapidly changing auditory information processing, it has been argued that enhancing basic musi...

The core deficit underlying developmental dyslexia (DD) has been identified in difficulties in dynamic and rapidly changing auditory information processing, which contribute to the development of impaired phonological representations for words. It has been argued that enhancing basic musical rhythm perception skills in children with DD may have a p...

Tile coding is an effective reinforcement learning method that uses a rather ingenious generalization mechanism based on (1) a carefully designed parameter setting and (2) the assumption that nearby states in the problem space will correspond to similar payoff predictions in the action-value function. Previously, we extended XCSF with tile coding p...

The XCS classifier system is a rule-based evolutionary machine learning system. XCS evolves classifiers in order to learn generalized solutions. The XCS with adaptive action mapping (XCSAM) is inherited from XCS, which evolves a best action map where it evolves classifiers that advocate the best action in every state. Accordingly, XCSAM can potent...

Learning Classifier Systems (LCS) are population-based reinforcement learners that were originally designed to model various cognitive phenomena. This paper presents an explicitly cognitive LCS by using spiking neural networks as classifiers, providing each classifier with a measure of temporal dynamism. We employ a constructivist model of growth o...

Learning Classifier Systems (LCS) are population-based reinforcement learners that were originally designed to model various cognitive phenomena. This paper presents an explicitly cognitive LCS by using spiking neural networks as classifiers, providing each classifier with a measure of temporal dynamism. We employ a constructivist model of growth o...

TrackGen is an online tool for the generation of tracks for two open-source 3D car racing games (TORCS and Speed Dreams). It integrates interactive evolution with procedural content generation and comprises two components: (i) a web frontend that maintains the database of all the evolved populations and manages the interaction with users (by collec...

A learning strategy in Learning Classifier Systems (LCSs) defines how classifiers cover a state-action space in a problem. Previous analyses in classification problems have empirically claimed an adequate learning strategy can be decided depending on types of noise. This issue is still arguable from two aspects. First, there lacks comparison of lea...

Exergames provide efficient and motivating training mechanisms to support physical rehabilitation at home. Nonetheless, current exergame examples lack some important aspects which cannot be disregarded in rehabilitation. Exergames should: (i) modify the game difficulty adapting to patient's gameplay performance, (ii) monitor if the exercise is corr...

The aim of this paper is to present a methodology for automatic tuning of a computational model, in the context of the development of flight simulators. We will present alternative approaches to automatic parameter ranking and screening, developed in a collaboration between Politecnico di Milano and TXT e-solutions and designed to fit as much as po...

Match balancing is one of the most important design issues in the development of an adversarial multiplayer shooter. Therefore, matchmaking algorithms are generally used to build teams, such that players can have fun with each other and enjoy the game experience. In this work we approach this problem from a completely different angle: we show that...

Artificial intelligence (AI) plays a major role in modern video games by making them feel both more realistic and more fun to play. Game intelligence usually works alongside the game logic, in the background, invisible to the players who enjoy the resulting character behaviors, the adaptive gameplay, and the procedurally generated content. However,...

We extend XCS with computed prediction by replacing the usual lin-ear prediction used in XCSF with a feedforward multilayer neural network. In XCSF with neural prediction, XCSFNN, classifier exploits a neural network to approximate the payoff surface associated to the target problem while the genetic algorithm adapts both classifier conditions, cla...

We study two existing Learning Classifier Systems (LCSs): XCS, which has a complete map (which covers all actions in each state), and XCSAM, which has a best action map (which covers only the highest-return action in each state). This allows XCSAM to learn with a smaller population size limit (but larger population size) and to learn faster than XC...

Learning Classifier Systems were introduced in the 1970s by John H. Holland as highly adaptive, cognitive systems. More than 40 years later, the introduction of Stewart W. Wilson's XCS, a highly engineered classifier system model, has transformed them into a state-of-the-art machine learning system. Learning classifier systems can effectively solve...

Rehabilitation is now more costly than ever and novel solutions are needed to cope with such increasing costs. Exergames at home can be a viable solution, allowing the patients to exercise in the comfort of their own home while playing simple games. However, to provide all the functionalities that a traditional therapy would, an adequate architectu...

Computer games are a promising tool to support intensive rehabilitation. However, at present, they do not incorporate the supervision provided by a real therapist and do not allow a safe and effective use at patient’s home. We show here how specifically tailored computational intelligence based techniques allow extending exergames with functionalit...

In car racing, blocking refers to maneuvers that can prevent, disturb or completely block an overtaking action by an incoming car. In this paper, we present an advanced overtaking behavior that is able to deal with opponents implementing blocking strategies of various difficulty level. The behavior we developed has been integrated in an existing fu...

Natural User Interfaces allow users to interact with virtual environments with little intermediation. Immersion becomes a vital need for such interfaces to be successful and it is achieved by making the interface invisible to the user. For cognitive rehabilitation, a mirror view is a good interface to the virtual world, but obtaining immersion is n...

We present an application of data mining to the analysis and tuning of Bad Blood, a video game for Windows Phone developed for the 2012 Microsoft's Imagine Cup competition. The game was developed on a very short time frame (four months) by a small student team (three programmers and one designer). Because of the limited development time available t...

Physical and cognitive rehabilitation under the form of therapeutic videogames has been growing in popularity over the last years. Many rehabilitation games (or exergames) have been created with the intent to promote functional rehabilitation in a highly motivational environment. However, such exergames are often created as standalone products typi...

XCS with Adaptive Action Mapping (XCSAM) evolves so- lutions focused on classifiers that advocate the best action in every state. Accordingly, XCSAM usually evolves more compact solutions than XCS which, in contrast, works to- ward solutions representing complete state-action mappings. Experimental results have however shown that, in some prob- lem...

Modern heterogeneous embedded platforms, composed of several digital signal, application specific and general purpose processors, also include reconfigurable devices supporting partial dynamic reconfiguration. These devices can change the behavior of some of their parts during execution, allowing hardware acceleration of more sections of the applic...

This paper proposes a novel rule discovery mechanism for the XCS classifier system, which is an extension of the compact genetic algorithm (cGA) to XCS. Our rule discovery mechanism, like cGA, extracts appropriate attributes of classifier conditions through a probability vector and evolves classifiers using the extracted attributes. Unlike cGA, it...

The recent availability of advanced video game interfaces (such as the Microsoft Kinect, the Nintendo WiiMote and Balance Board) is creating interesting opportunities to provide low-cost rehabilitation at-home for patients. In this context, video games are rising as promising tools to guide patients through their recovery experience and to increase...

This manual describes the competition software for the Simulated Car Racing Championship, an international competition held at major conferences in the field of Evolutionary Computation and in the field of Computational Intelligence and Games. It provides an overview of the architecture, the instructions to install the software and to run the simpl...

The aim of this article is to describe a game engine that has all the characteristics needed to support rehabilitation at home. The low-cost tracking devices recently introduced in the entertainment market allow measuring reliably at home, in real time, players' motion with a hands-free approach. Such systems have also become a source of inspiratio...

Modern satellite telecommunications are moving toward the use of large capacity systems exploiting fade mitigation techniques such as reconfigurable on-board antenna systems to face channel attenuation. In this work we propose a new Single-Objective Genetic Algorithm (SOGA) able to optimize the spatial distribution of the radiated power, taking int...

Exergames for rehabilitation, both in the physical and cognitive fields, have been the target of much research in the last years. Such exergames, however, are often created for a specific impairment and cannot be generalized to other domains. More generally speaking, the lack of shared design and development guidelines for rehabilitation games can...

The XCS classifier system evolves solutions that represent complete mappings from state-action pairs to expected returns therefore, in every possible situation, XCS can predict the value of all the available actions. Such complete mapping is sometimes considered redundant as most of the applications (like for instance, classification), usually focu...

We show here how integrating novel natural user interfaces, like Microsoft Kinect, with a fully adappatient's current statustive game engine, a system that can be used for rehabilitation at home can be built. A wide variety of game scenarios, a balanced scoring system, quantitative and qualitative exercise evaluation, automatic gameplay level adapt...

Finding a racing line that allows to achieve a competitive lap-time is a key problem in real-world car racing as well as in the development of non-player characters for a commercial racing game. Unfortunately, solving this problem generally requires a domain expert and a trial-and-error process. In this work, we show how evolutionary computation ca...

Computer games are a promising tool to support rehabilitation at home. It is widely recognized that rehabilitation games should (i) be nicely integrated in general-purpose rehabilitation stations, (ii) adhere to the constraints posed by the clinical protocols, (iii) involve movements that are functional to reach the rehabilitation goal, and (iv) ad...

This paper proposes a novel approach of XCS called XCS with Best Action Mapping (XCSB) to enhance the learning capabilities of XCS. The feature of XCSB is to learn only best actions having the highest predicted payoff with the high accuracy unlike XCS which learns actions having the highest and lowest predicted payoff with the high accuracy. To inv...

Geo-Demographic Analysis (GDA) is an important tool to explore the underlying rules that regulate our world, and therefore, it has been widely applied to the development of effective socio-economic policies through the analysis of data generated from Geographic Information Systems (GIS). In GDA applications, clustering plays a major role however, t...

An automated technique has recently been proposed to transfer learning in the hierarchical Bayesian optimization algorithm (hBOA) based on distance-based statistics. The technique enables practitioners to improve hBOA efficiency by collecting statistics from probabilistic models obtained in previous hBOA runs and using the obtained statistics to bi...

Learning Classifier Systems (LCS) are population-based reinforcement learners used in a wide variety of applications. This paper presents a LCS where each traditional rule is represented by a spiking neural network, a type of network with dynamic internal state. We employ a constructivist model of growth of both neurons and dendrites that realise f...

Geographic Information Systems (GIS) play a very important role to researches and industries. In fact, there have been some studies related to the development of this kind of systems as well as methods of mining attribute GIS data. In this paper, we will explore an aspect of Data Mining in GIS, that is, GIS Clustering for Geo-Demographic Analysis a...

Transfer learning might be a promising approach to boost the learning of non-player characters' behaviors by exploiting some existing knowledge available from a different game. In this paper, we investigate how to transfer driving behaviors from The Open Racing Car Simulator (TORCS) to VDrift, which are two well known open-source racing games featu...

Cheating reportedly affects most of the multi-player online games and might easily jeopardize the game experience by providing an unfair competitive advantage to one player over the others. Accordingly, several efforts have been made in the past years to find reliable and scalable approaches to solve this problem. Unfortunately, cheating behaviors...

In this paper, we investigate the application of evolutionary computation to the automatic generation of tracks for high-end racing games. The idea underlying our approach is that diversity is a major source of challenge/interest for racing tracks and, eventually, might play a key role in contributing to the player's fun. In particular, we focus on...

We present a framework for the procedural generation of tracks for a high-end car racing game (TORCS) using interactive evolution. The framework maintains multiple populations and allow users to work both on their own population (in single-user mode) or to collaborate with other users on a shared population. Our architecture comprises a web fronten...

We address the problem of automatically designing maps for first-person shooter (FPS) games. An efficient solution to this procedural content generation (PCG) problem could allow the design of FPS games of lower development cost with near-infinite replay value and capability to adapt to the skills and preferences of individual players. We propose a...

Recently, the Terrain Splitting and Mapping problem [11] has been considered as one of some current striking trends in GIS researches. This problem is very important in making 3D WebGIS systems and handling large terrain data. In fact, these have been some algorithms dealing with it based on Conditional partitioning approaches [11], Greedy strategi...