You are here

Research topics

Il programma di ricerca della Scuola IMT è caratterizzato dalla complementarietà di metodologie di ambiti quali economia, neuroscienze cognitive e sociali, visual studies, filosofia, archeologia, storia dell’arte, diritto del patrimonio culturale, analisi e gestione del patrimonio culturale, informatica e ingegneria. Le descrizioni di alcuni esempi di progetti e di temi di ricerca attivi alla Scuola sono disponibili qui di seguito.

The IMT School's research program is characterized by the complementarity of methodologies from areas such as economics, cognitive and social neuroscience, visual studies, philosophy, archeology, art history, cultural heritage law, analysis and management of cultural heritage, computer science and engineering. The descriptions of some examples of projects and research themes active at the School are available below.

 

CCSN PROJECTS

CSSE PROJECTS

High performance computing for nonlinear coupled problems in solid and fluid mechanics
Abstract
Nonlinear coupled problems governed by partial differential equations in solid and fluid mechanics arise in many engineering and biological applications where multiple fields (displacement, damage, thermal, humidity, electric, etc.) are strongly interacting with each other. The present research topic envisages a critical analysis and development of novel numerical strategies for the solution of nonlinearly coupled boundary value problems within the finite element method. Specifically, implicit and explicit numerical schemes, as well as monolithic and staggered solvers, along with suitable high performance computing strategies, will be developed for a wide range of problems selected for their relevance in industrial applications and failure analysis. Prospective applicants are expected to hold a degree in engineering, mathematics, physics, or computer science.
Keywords
nonlinear partial differential equations; coupled problems; high performance computing; solid mechanics; fluid mechanics.
References
  • Reinoso J, Paggi M, Linder C (2017). Phase field modeling of brittle fracture for enhanced assumed strain shells at large deformations: formulation and finite element implementation. COMPUTATIONAL MECHANICS, vol. 59, p. 981-1001, doi: 10.1007/s00466-017-1386-3
  • Lenarda P, Paggi M, Ruiz Baier R (2017). Partitioned coupling of advection–diffusion–reaction systems and Brinkman flows. JOURNAL OF COMPUTATIONAL PHYSICS, vol. 344, p. 281-302, doi: 10.1016/j.jcp.2017.05.011
  • Lenarda P, Gizzi A, Paggi M (2018). A modeling framework for electro-mechanical interaction between excitable deformable cells. EUROPEAN JOURNAL OF MECHANICS. A, SOLIDS, vol. 72, p. 374-392, doi: 10.1016/j.euromechsol.2018.06.001

Reference Faculty
Marco Paggi (MUSAM) and Mirco Tribastone (SYSMA)

Adhesive and cohesive failures in structural adhesives: the interplay between chemistry and mechanics
Abstract
Structural adhesives are used in many industrial applications and are currently designed to guarantee a prescribed load carrying capacity and optimal sealing of the joint. Failures of such joints can be either cohesive or adhesive. In the former case, the crack pattern takes place across the adhesive material, which has its own thickness. In the latter, the interface between the adhesive and the substrate is the weakest link and it leads to premature delamination. In many intermediate situations, both failure modes are concurrently observed. This research topic aims at fully characterizing such failure modes and at understanding how chemical surface treatments can affect the mechanical response of the joint. Both experimental tests in the MUSAM-Lab and numerical research by exploiting the capabilities of the novel phase-field formulation for fracture coupled with the cohesive zone model for delamination will be conducted. Prospective applicants are expected to hold a degree in applied chemistry, materials science, engineering, physics or mathematics.
Keywords
adhesive and cohesive failures; chemical surface treatments; laboratory testing; computational modelling.
References
  • Reinoso J, Paggi M (2014). A consistent interface element formulation for geometrical and material nonlinearities. COMPUTATIONAL MECHANICS, vol. 54, p. 1569-1581, doi: 10.1007/s00466-014-1077-2
  • Paggi M, Reinoso J (2017). Revisiting the problem of a crack impinging on an interface: A modeling framework for the interaction between the phase field approach for brittle fracture and the interface cohesive zone model. COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING, vol. 321, p. 145-172, doi: 10.1016/j.cma.2017.04.004
  • Mariggiò G, Reinoso J, Paggi M, Corrado M (2018). Peeling of thick adhesive interfaces: The role of dynamics and geometrical nonlinearity. MECHANICS RESEARCH COMMUNICATIONS, vol. 94, p. 21-27, doi: 10.1016/j.mechrescom.2018.08.018
Reference Faculty
Marco Paggi (MUSAM)
Contact mechanics between rough surfaces: advanced computational modelling and simulation
Abstract
Roughness plays a key role in surface phenomena such as surface physics (heat and electric transfer, hydrophobicity, etc.), surface chemistry (chemical reactions, diffusion, etc.) and tribology (stress transfer, adhesion, lubrication, etc.). Frontier research topics regard the development of finite element-based computational methods allowing for the simulation of contact problems with multiple fields and nonlinear constitutive relations, taking also into account the emergent behaviour induced by microscopic surface roughness. The present research will exploit the new MPJR finite element framework recently published by Paggi and Reinoso, further extending it to rough surfaces in tangential contact and under the action of multiple fields. Joint co-supervision with Prof. Reinoso will be proposed, allowing for the appointment of a double PhD degree at IMT and at the University of Seville, Spain.
Prospective applicants are expected to hold a degree in engineering, physics or mathematics.
Keywords
contact mechanics; roughness; finite element method; coupled problems; nonlinear constitutive relations.
References
  • Paggi M, Barber JR (2011). Contact conductance of rough surfaces composed of modified RMD patches. INTERNATIONAL JOURNAL OF HEAT AND MASS TRANSFER, vol. 54, p. 4664-4672, doi:10.1016/j.ijheatmasstransfer.2011.06.011
  • Vakis AI, Yastrebov VA, Scheibert J, Nicola L, Dini D, Minfray C, Almqvist A, Paggi M, Lee S, Limbert G, Molinari JF, Anciaux G, Aghababaei R, Echeverri Restrepo S, Papangelo A, Cammarata A, Nicolini P, Putignano C, Carbone G, Stupkiewicz S, Lengiewicz J, Costagliola G, Bosia F, Guarino R, Pugno NM, Müser MH, Ciavarella M (2018). Modeling and simulation in tribology across scales: An overview. TRIBOLOGY INTERNATIONAL, vol. 125, p. 169-199, doi: 10.1016/j.triboint.2018.02.005
  • Paggi M, Reinoso J (2018). A variational approach with embedded roughness for adhesive contact problems, MECHANICS OF ADVANCED MATERIALS AND STRUCTURES, in press, doi:10.1080/15376494.2018.1525454
Reference Faculty
Marco Paggi (MUSAM)
Optimization of additive manufacturing solutions for higher reliability and durability of composites
Abstract
Additive manufacturing solutions are enabling a new era of design optimization, complexity and functionality for composite structures. With the advent of 3D printing technologies, additive manufacturing solutions have rapidly advanced and reached a state of mainstream adoption, particularly for rapid prototyping. Such technologies are only beginning to penetrate and influence the advanced composites industry. The present research project aims at realizing a comprehensive analysis and a critical comparison of the existing additive manufacturing solutions, with special attention to their specific processes. Research will focus on the issues of reliability and durability of composite components realized by such techniques, exploiting computational mechanics tools to simulate each manufacturing process. Optimization strategies will be also explored in order to improve geometries, material combinations, and process parameters towards maximizing the mechanical performance of composites and their durability. Prospective applicants are expected to hold a degree in engineering or mathematics.
Keywords
additive manufacturing solutions; composites; computational mechanics; computational optimization.
Reference Faculty
Marco Paggi (MUSAM) and Alberto Bemporad (DYSCO)
Modeling and control of software performance
Abstract
Response time, throughput and utilization are extra-functional properties of software that are especially relevant in resource-constrained environments such as mobile phones and in the Internet-of-Things. Ideally, one would like to use an application that automatically meets given user-defined performance requirements. This requires the availability of a mechanism that can identify the current state of the software system and predict its future behaviour under a range of assumptions of the environment, with an algorithm that returns the optimal configuration meeting the desired performance target. The candidate will have the opportunity to work on the development of self-adaptive methods for software performance using a range of techniques including black-box representations based on machine learning and white-box analytical models built from first principles. 
Keywords
software performance engineering; self-adaptive systems; predictive modelling
Reference Faculty
Mirco Tribastone
References
E. Incerto, M. Tribastone, and C. Trubiani, “Software performance self-adaptation through efficient model predictive control,” in 32nd ACM/IEEE International Conference on Automated Software Engineering (ASE), 2017. http://cse.lab.imtlucca.it/∼mirco.tribastone/papers/ase2017.pdf 
E. Incerto, M. Tribastone, and C. Trubiani, “Combined vertical and horizontal autoscaling through model predictive control,” in 24th International European Conference on Parallel and Distributed Computing (EURO-PAR), 2018. http://cse.lab.imtlucca.it/∼mirco.tribastone/papers/europar18.pdf 
E. Incerto, A. Napolitano, and M. Tribastone, “Moving horizon estimation of service demands in queuing networks,” in 26th IEEE International Symposium on the Modeling, Analysis, and Simulation of Computer and Telecommunication Systems (MASCOTS), 2018. http://cse.lab.imtlucca.it/∼mirco.tribastone/papers/mascots18.pdf 
Coarse-graining of dynamical systems

Abstract
Dynamical systems are a fundamental mathematical model to describe predict the behavior of natural as well as engineered processes. Our capability to gain relevant insights details is however hindered by the large dimensionality of such models when describing systems characterized by a large degree of complexity. Coarse graining, model reduction, and model abstraction are among the several different keywords with which a wide range of disciplines refer to the topic of simplifying a given dynamical system into a smaller one that preserves key observables of interest to the modeler. The candidate will have the opportunity to work on the development of new coarse-graining methods and algorithms, with applications related to models in various disciplines including automation, computer science, statistical physics, and systems biology.
Keywords
coarse graining; model reduction; dynamical systems
Reference Faculty
Mirco Tribastone (SYSMA), Guido Caldarelli (NETWORKS), and Diego Garlaschelli (NETWORKS)
References

  • S. Tognazzi, M. Tribastone, M. Tschaikowski, and A. Vandin, “Backward invariance for linear differential algebraic equations,” in 57th IEEE Conference on Decision and Control (CDC), 2018
  • L. Cardelli, M. Tribastone, M. Tschaikowski, and A. Vandin, “Maximal aggregation of polynomial dynamical systems,” Proceedings of the National Academy of Sciences, vol. 114, no. 38, pp. 10 029–10 034, 2017.
  • L. Cardelli, M. Tribastone, A. Vandin, and M. Tschaikowski, “ERODE: A tool for the evaluation and reduction of ordinary differential equations,” in Tools and Algorithms for the Construction and Analysis of Systems - 23rd International Conference, TACAS, 2017.
Learning and identification for control

Abstract
The goal is to develop autonomous, self-reconfigurable, control systems that are able to learn how to achieve their objectives from data, adapting themselves to external stimuli such as changing environmental conditions and variations of the dynamic properties of the process. In particular, new approaches will be developed for synthesizing control systems from data that are optimal, robust, and can cope with operating constraints on input and output variables, addressing both model-based methods, where an open-loop model of the process is identified from data, and model-free methods, that directly synthesize the control law from data.
Keywords
Systems identification, machine learning, reinforcement learning, model predictive control
Reference Faculty
A. Bemporad
References

  • R. S. Sutton and A. G. Barto, Reinforcement learning: An introduction. Second Edition. MIT press Cambridge, 2018
  • F. L. Lewis, D. Vrabie, and K. G. Vamvoudakis, “Reinforcement learning and feedback control: Using natural decision methods to design optimal adaptive controllers,” IEEE Control Systems, vol. 32, no. 6, pp. 76–105, 2012.
  • B. Recht, "A Tour of Reinforcement Learning: The View from Continuous Control", 2018.
  • D. Selvi, D. Piga, and A. Bemporad, “Towards direct data-driven control design of optimal controllers,” in Proc. European Control Conf., Limassol, Cyprus, 2018, pp. 2836–2841.
  • V. Breschi, D. Piga, and A. Bemporad, “Piecewise affine regression via recursive multiple least squares and multicategory discrimination,” Automatica, vol. 73, pp. 155–162, Nov. 2016.
  • D. Piga, S. Formentin, and A. Bemporad, “Direct data-driven control of constrained systems,” IEEE Transactions on Control Systems Technology, vol. 26, no. 4, pp. 1422–1429, Jul 2018.
  • J. R. Salvador, D. Munoz de la Pena, T. Alamo, and A. Bemporad, “Data-based predictive control via direct weight optimization,” in 6th IFAC Conference on Nonlinear Model Predictive Control, pp. 437-442, Madison, WI, USA, 2018.
  • D. Masti and A. Bemporad, “Learning nonlinear state-space models using deep autoencoders,” in Proc. 57th IEEE Conf. on Decision and Control. 2018./
MPC for autonomous driving / robotics

Abstract
In recent years, robots have spread everywhere and are able to accomplish numerous tasks in various applications. While most success has been obtained in restricted environments, the current challenges include being able to operate safely (e.g. not harming people around the robots), coordinating multiple robots and adapting to changing environments.In this context, Model Predictive Control is a valuable tool for tackling constrained multiple input multiple output (possibly nonlinear) systems. While both theory and algorithms have been largely investigated in the literature, many questions still need to be answered before it will be possible to safely and effectively deploy robots to cooperate with human beings. Autonomous driving is currently perhaps the most thrilling field of research in this context.
Keywords
Robotics, autonomous driving, model predictive control
Reference Faculty
M. Zanon
References

  • Craig, J. J. (2005). Introduction to robotics: mechanics and control (Vol. 3, pp. 48-70). Upper Saddle River, NJ, USA:: Pearson/Prentice Hall.
  • Borrelli, F., Bemporad, A., & Morari, M. (2017). Predictive control for linear and hybrid systems. Cambridge University Press.- Rawlings, J. B., & Mayne, D. Q. (2009). Model predictive control: Theory and design.
  • M. Graf Plessen, D. Bernardini, H. Esen, and A. Bemporad. Spatial-based predictive control and geometric corridor planning for adaptive cruise control coupled with obstacle avoidance. IEEE Trans. Contr. Systems Technology, vol. 26, no. 1, pp. 38–50, 2018.
  • R. Hult, M. Zanon, S. Gros, and P. Falcone. Optimal Coordination of Automated Vehicles at Intersections: Theory and Experiments. IEEE Transactions on Control Systems Technology, (in press, available online)
  • I. Batkovic, M. Zanon, N. Lubbe and P. Falcone. A Computationally Efficient Model for Pedestrian Motion Prediction. Proceedings of the European Control Conference (ECC), 2018
Nonconvex distributed optimization

Abstract
Distributed Optimization has been mostly investigated in order to address problems which have an intrinsically distributed nature, including smart cities, autonomous driving, power grids, etc. By restricting the focus on convex problems, it is possible to apply a wide range of algorithms. Moreover, the assumption of lossless and instantaneous communication links makes it possible to deploy algorithms which require simple computations but rely on large amounts of communication. In order to extend the applicability of distributed optimization, investigating nonconvex optimization strategies over lossy communication channels becomes of paramount importance. While only few results are currently available for this setting, some encouraging results have recently been obtained, which pave the road for further investigations on the topic.  Notable examples include networked and cooperative control scenarios applied to, e.g., autonomous driving, formation flight, etc.
Keywords
Nonconvex optimization, distributed optimization, networked optimization
Reference Faculty
M. Zanon
References

  • S. Boyd, L. Vandenberghe, "Convex optimization", Cambridge Univ. Press, 2004
  • J. Nocedal, S.J. Wright, Numerical optimization 2nd ed, 2006
  • R. Hult, M. Zanon, S. Gros and P. Falcone. Primal Decomposition of the Optimal Coordination of Vehicles at Traffic Intersections.Proceedings of the Conference on Decision and Control (CDC), 2016
  • M. Zanon, S. Gros, P. Falcone and H. Wymeersch. An Asynchronous Algorithm for Optimal Vehicle Coordination at Traffic Intersections. Proceedings of the World Congress of the International Federation of Automatic Control, 2017
Numerical methods for model predictive control

Abstract
Model Predictive Control (MPC) is one of the most successful techniques adopted in industry to control multivariable systems in an optimized way under constraints on input and output variables. In MPC, the manipulated inputs are computed in real time by solving a mathematical programming problem, most frequently a Quadratic Program (QP). Topics of research are available on how to formulate and solve MPC problems with higher throughput than is currently available, while maintaining the numerical optimization code simple, certifiable for the worst-case execution time, and robust with respect to limited machine precision.
Keywords
Convex optimization, numerical methods, model predictive control
Reference Faculty
A. Bemporad
References

  • G. Cimini and A. Bemporad, "Exact complexity certification of active-set methods for quadratic programming," IEEE Trans. Automatic Control, vol. 62, no. 12, pp. 6094–6109, 2017.
  • A. Bemporad, "A numerically stable solver for positive semi-definite quadratic programs based on nonnegative least squares," IEEE Trans. Automatic Control, vol. 63, no. 2, pp. 525–531, 2018.
  • A. Bemporad and V.V. Naik, "A numerically robust mixed-integer quadratic programming solver for embedded hybrid model predictive control," in 6th IFAC Conf. on Nonlinear Model Predictive Control, Madison, WI, 2018, pp. 502–507.
  • N. Saraf and A. Bemporad, "Fast model predictive control based on linear input/output models and bounded-variable least squares," in Proc. 56th IEEE Conf. on Decision and Control, Melbourne, Australia, 2017.
  • A. Bemporad, M. Morari, V. Dua, and E.N. Pistikopoulos, "The explicit linear quadratic regulator for constrained systems," Automatica, vol. 38, no. 1, pp. 3–20, 2002.
  • J. Nocedal, S.J. Wright, Numerical optimization 2nd ed, 2006
  • R. Verschueren, M. Zanon, R. Quirynen and M. Diehl. Time-optimal Race Car Driving using an Online Exact Hessian based Nonlinear MPC Algorithm. Proceedings of the European Control Conference (ECC), 2016
VeriOSS: A Blockchain for Open Source Software Verification

Abstract
The Software Supply Chain (SSC) is a cornerstone of the industrial society on which many other Supply Chains (SCs) depend. The continuous demand/integration of the computing systems into SCs is pushing the development and distribution of software. To cope with this growing request many companies are including open source software (OSS) in their software products. OSS has many advantages, for example, it prevents that the software producer does not acquires a strong bargaining position on the consumer. However, the flip side is that the producer of a OSS has no obligation to maintain, improve or fix her software. All in all, the OSS ranges from small scale projects, with limited or even no security plan, to community projects that release periodic security updates. Such heterogeneity makes it difficult to understand the actual risks when one wants to integrate a OSS in his project. From a methodological point of view, the project aims at answering the following questions: (i) what are the conditions to make the formal verification a valuable asset in the SSC? (ii) can we design a mechanism based on economic rewards that push participant to find and fix bugs in OSS software? (iii) can the blockchain technology be used to implement a decentralized framework for the formal verification of security properties of OSS? From a practical point of view, the project aims at designing and implementing a blockchain service for the security analysis and patching of the OSS, where developers and security analysts cooperate efficiently.
Keywords
Formal methods, software verification, blockchain, DLT, contract-driven development, mobile code security
Reference Faculty
Letterio Galletta
Research units involved
SysMA, AXES

Automatic vulnerability disclosure with coevolutionary algorithms

Abstract
Symbolic execution is a powerful technique to spot out corner cases, e.g., vulnerabilities, in the semantics of a program. As a matter of fact, it replaces the standard semantics (referring to specific values) with a symbolic one (manipulating abstract expressions). Unfortunately, symbolic execution does not scale on large programs. For this reason, hybrid techniques have been proposed (e.g. concolic testing and symbolic backward execution). The goal of this project is to combine the symbolic analysis of a program with a test execution environment driven by an evolutionary optimization strategy. The symbolic analysis is applied to obtain a compact (thus computable) specification of the conditions that a test must satisfy to trigger a certain vulnerability. Instead the evolutionary algorithm drives the refinement process that, starting from some random tests, leads to the convergence toward the desired ones. The convergence criteria is based on the optimization of a fitness function derived from the symbolic specification.
References
Formal methods, vulnerability analysis, security testing, evolutionary algorithms, white-box testing, binary analysis
Reference Faculty
Gabriele Costa
Research units involved
SysMA, AXES

ENBA PROJECTS

Scale-invariance in Networks, from Renormalization Group to Aggregation. Theory and application to Biochemical reactions.
Abstract
The property of scale-invariance of a physical system allows to determine new and universal properties in the evolution of the system itself. For example, in the evaluation of the scaling properties of the system Hamiltonian it is possible to determine analytically the value of the critical exponents. Graphs are particularly difficult problem to approach since the small-world effect that limits the decimation procedure to few steps determined by the diameter of the instance available. Nonetheless the problem has been approached already in some limit cases, since the possible applications are incredibly important spanning from computer science to statistical physics and to ecology (trophic species). The candidate will work on theoretical models (spanning trees, cayley trees) and applications from biochemistry and social systems.
References
  • Eigenvector centrality for characterization of protein allosteric pathways
  • Christian F. A. Negre, PNAS 115 (52) E12201-E12208 (2018)
  • Cardelli, L., Tribastone, M., Tschaikowski, M., and Vandin, A. (2015). Forward and backward bisimulations for chemical reaction networks. arXiv preprint arXiv:1507.00163.
  • Cardelli, L., Tribastone, M., Tschaikowski, M., and Vandin, A. (2017). Maximal aggregation of polynomial dynamical systems. Proceedings of the National Academy of Sciences, 114(38):10029–10034.
  • J. Cardy Scaling and Renormalization in Statistical Physics OUP (1996)
Reference Faculty
Guido Caldarelli, Diego Garlaschelli, Mirco Tribastone
Fake news propagation and cybersecurity: Diffusion in Disordered media
Abstract
The issue of fake news (and the problems arising from them) recently became of great importance for the society. Indeed, while fake news and propaganda always existed in the history of mankind, the computer revolution that created the present connected world made the problem more and more pressing. This happened because it is now possible for everybody to have access to a global audience and to be able to address political/economic/religious and all the other social issues with a simple access to the Internet. Internet and the network of social contacts of Internet users are distributed with a particularly variate topology so that the study of news propagation and modification is particularly troublesome. The project of this thesis should model the propagation of news and their modification on a specific network, namely that of the Facebook users.
References
  • Hall, G. & Bialek, W. The statistical mechanics of Twitter. arXiv (2018).
  • Weng, L., Flammini, A., Vespignani, A. and Menczer, F. [2012], ‘Competition among memes in a world with limited attention’,Scientific Reports2, 1–9.11
Reference Faculty
Guido Caldarelli, Rocco De Nicola, and Fabrizio Silvestri (Facebook UK)
Extension of Fitness model to Epidemics
Abstract
Fitness Model of network formation is one simple statistical model that reproduces the property of scale-invariance of the real data. So far it has not been applied to fast varying networks where the properties of the vertices vary in time as in the case of epidemics. The work of the thesis will start from analytical derivations of model extensions to the validation of the computer result with respect to real data of epidemics.
References
  • Caldarelli, G., Capocci, A., De Los Rios, P. & Muñoz, M. A. Scale-Free Networks from Varying Vertex Intrinsic Fitness. Phys. Rev. Lett. 89, (2002).
  • Pastor-Satorras, R. & Vespignani, A. Epidemic Spreading in Scale-Free Networks. Phys. Rev. Lett. 86, 3200–3203 (2001).
  • Tizzoni, M., Sun, K., Benusiglio, D., Karsai, M. & Perra, N. The Scaling of Human Contacts and Epidemic Processes in Metapopulation Networks. Sci. Rep. 5, 15111 (2015).
Reference Faculty
Guido Caldarelli and Daniela Paolotti (ISI Turin)
Study of Country production from Leontief matrices
Abstract
We shall study the network properties of input-output matrices that is of the interrelation of the production and use of different economical sector within one country and across different countries. The candidate will apply the ideas of the network analysis to country exports to this specific case.
References
  • Leontief, Wassily W. Input-Output Economics. 2nd ed., New York: Oxford University Press, 1986.
  • The Product Space Conditions the Development of Nations. C. A. Hidalgo, B. Klinger, A.-L. Barabási, R. Hausmann 317 482-487 (2007)
Reference Faculty
Guido Caldarelli, Massimo Riccaboni and Armando Rungi
Study of the Network of Small Enterprises in Tuscany
Abstract
The candidate will collect and analyse data on the network of small enterprises in Tuscany in order to determine the topology of contacts between the different companies and possibly in order to establish the stability of this system with respect to credit risk. The candidate will also work analytically on models of credit to enterprises.
References
  • Complex Agent Based Models M. Gallegati Springer 2018
  • The financial accelerator in an evolving credit network. Domenico Delli Gatti, Mauro Gallegati, Bruce Greenwald, Alberto Russo, Joseph E. Stiglitz. Journal of Economic Dynamics & Control 34 (2010) 1627–1650
Reference Faculty
Guido Caldarelli, Nicola Lattanzi and Fabio Saracco
Prosociality, Cognition and Peer Effects
Abstract
This line of research investigates the relationship between the mode of cognition, the structure of social interactions with peers and the likelihood of prosocial behavior. The fundamental drivers of prosocial behavior, i.e., the "social motives", are likely to be fairly stable over time, since they are mainly shaped by social interactions and peer effects. However, which of the social motives actually has the greatest role in shaping behavior in a given situation is likely to depend on the mode of cognition adopted in that situation by the decision-maker. To investigate this we need both theoretical analysis (with models and simulations) and empirical analysis (with experiments and survey data) in order to explore the role of: attention, cognitive effort, empathy, moral reasoning, intrinsic motivations, experienced and expected social sanctioning. The results of this research line should help us better understand the determinants of prosocial behavior and, in particular, to derive implications for policies designed to foster prosociality.
Keywords
Cooperation, Intuition and deliberation, Reciprocity, Social sanctioning, Social network
References
  • Thöni, Christian, and Simon Gächter. "Peer effects and social preferences in voluntary cooperation: A theoretical and experimental analysis." Journal of Economic Psychology 48 (2015): 72-88.
  • Belloc, Marianna, Ennio Bilancini, Leonardo Boncinelli, and Simone D’Alessandro. "Intuition and Deliberation in the Stag Hunt Game" (2019), mimeo
  • Bilancini, Ennio, and Leonardo Boncinelli. "The co-evolution of cooperation and defection under local interaction and endogenous network formation." Journal of Economic Behavior & Organization 70.1-2 (2009): 186-195.
Main researcher
Ennio Bilancini (AXES)
Research units involved
AXES, MOMILAB, NETWORKS, SYSMA
Evolution of Social Behaviors: Norms and Institutions
Abstract
Social norms are patterns of behavior that are self-enforcing within a group: if everyone is expected to conform to the ruling social norm, then everyone actually wants to conform. Social norms are often sustained by different mechanisms: desire to coordinate, fear of being sanctioned, signaling membership in a group, or simply following a leader. The rule of 50-50 sharecropping is an example of a social norm. Institutions are explicit rules (formal or informal) that are enforced by the behaviors and social norms of a society. Social status is a prominent example of an informal institution, while a caste system is an example of a formal one. Stochastic evolutionary game theory can be applied to study the emergence of social norms and institutions. Also, simulations of complex evolutionary dynamics and agent-based modeling are valuable tools in this regard. Further, the instrumental approach to social preferences can shed light on the evolutionary roots of social norms and institutions.
Keywords
Assortativity, Evolutionary dynamics, Evolutionary game theory, Learning, Stochastic stability
References
  • Bilancini, Ennio, Leonardo Boncinelli, and Jiabin Wu. "The interplay of cultural intolerance and action-assortativity for the emergence of cooperation and homophily." European Economic Review 102 (2018): 1-18.
  • Bilancini, Ennio, and Leonardo Boncinelli. "Social coordination with locally observable types." Economic Theory 65.4 (2018): 975-1009.
  • Bilancini, Ennio, and Leonardo Boncinelli. "Instrumental cardinal concerns for social status in two-sided matching with non-transferable utility." European Economic Review 67 (2014): 174-189.
Main researcher
Ennio Bilancini (AXES)
Research units involved
AXES, NETWORKS, SYSMA
Neuroeconomics of Strategic Interaction
Abstract
Game theory extends the model of an individual decision-maker to the case of multiple interacting decision-makers. Solution concepts predict which action profile will result from the actual playing of a game. The most prominent solution concept is Nash equilibrium which typically requires players to use rationality-based inference, common knowledge of beliefs and rationality. Such kind of reasoning can be cognitively extremely demanding and in many cases implausible. In particular, different levels of recursive thinking (i.e., a player’s mental processing that incorporates thinking about others) are likely to require substantial cognitive effort. Moreover, a growing body of experimental evidence reports frequent non-equilibrium play. Investigating which brain circuits are involved in strategic decision-making can help understand the roots of non-equilibrium play. Actually, neuroscientific evidence suggests that different portions of the prefrontal cortex distinguish high versus low levels of recursive thinking, as well as naive versus sophisticated learning, thus encoding the extent of strategic thinking. The aim of this research line is to develop novel models of strategic decision-making that are consistent with such evidence and to test experimentally both novel and existing models, providing new behavioral and neuroscientific data.
Keywords
Brain correlates of strategic reasoning, Bounded cognition, Game theory, Strategic sophistication, Recursive thinking
References
  • Alós-Ferrer, Carlos. "A Review Essay on Social Neuroscience: Can Research on the Social Brain and Economics Inform Each Other?." Journal of Economic Literature 56.1 (2018): 234-64.
  • Griessinger, Thibaud, and Giorgio Coricelli. "The neuroeconomics of strategic interaction." Current Opinion in Behavioral Sciences 3 (2015): 73-79.
  • Bilancini, Ennio, and Leonardo Boncinelli. "Rational attitude change by reference cues when information elaboration requires effort." Journal of Economic Psychology 65 (2018): 90-107.
Main researcher
Ennio Bilancini (AXES)
Research units involved
AXES, MOMILAB, NETWORKS
Measurement of Strategic Ability
Abstract
Strategic interactions have been studied extensively but so far no specific measure of a decision-maker's strategic ability has gained consensus. This research line aims at developing a framework that allows to measure strategic ability. Convincing conceptualizations of (bounded) rationality and cognitive effort as well as of mentalization (i.e., the construction of beliefs about others’ behavior) are crucial ingredients in the development of such framework. A good measure of the strategic ability of an individual should provide information on the likelihood of success that such an individual typically obtains in activities that involve strategic interaction among multiple decision-makers. To this aim original behavioral data have been extensively collected to construct a dataset of actual strategic behavior in a variety of strategic settings, together with socio-demographic and psychological measures. The analysis of such rich dataset should provide a test for the proposed measures of strategic ability as well as source of inspiration for developing novel behavioral measures.
Keywords
Depth of reasoning, Strategic Quotient, Behavioral game theory, Rationality, Theory of Mind
References
  • Bilancini, Ennio, Leonardo Boncinelli, and Alan Mattiassi. "Assessing Actual Strategic Behavior to Construct a Measure of Strategic Ability." Frontiers in Psychology (2019) forthcoming
  • Gill, David, and Victoria Prowse. "Cognitive ability, character skills, and learning to play equilibrium: A level-k analysis." Journal of Political Economy 124.6 (2016): 1619-1676.
  • Alaoui, Larbi, and Antonio Penta. "Endogenous depth of reasoning." Review of Economic Studies 83.4 (2015): 1297-1333.
Main researcher: Ennio Bilancini (AXES)
Research units involved: AXES, MOMILAB
Behavioral Economics of Health and Wellbeing
Abstract
This line of research investigates how we can improve health and well-being of citizens and populations acting on the organization of health care systems, behaviors of health care professionals, habits of individuals and communities. To this aim, methods and tools of behavioral and experimental economics are paired with expertise from life and health sciences, ergonomics and human factors, quality of life studies.
Keywords
Behavioral health economics, Health incentives, Quality and Safety of health care, Subjective well-being, Support to decision-making
References
  • Rice, Thomas. "The behavioral economics of health and health care." Annual review of public health 34 (2013): 431-447.
  • Bellandi, Tommaso, Sara Albolino, Riccardo Tartaglia, and Sebastiano Bagnara. "Human factors and ergonomics in patient safety management." In Handbook of Human Factors and Ergonomics in Health Care and Patient Safety, pp. 698-717. CRC Press, 2016.
  • Bartolini, Stefano, Ennio Bilancini, Luigino Bruni, and Pierluigi Porta, eds. Policies for happiness. Oxford University Press, 2016.
Main researcher
Ennio Bilancini
Research units involved
AXES
Technological change, soft skills and the future of high skilled jobs
Abstract
Advances in information technology and in the globalization of production have been structurally changing labor markets, posing major new challenges to policy makers. Take the case of the “jobless recovery” after the Great Recession, which has been at the center of a lively debate on the possible remedies to structural unemployment. In fact, recent findings suggest a stagnating job market for high skilled professionals since the 2000s driven by a decline in science, technology, engineering and math (STEM) occupations (Beaudry et al., 2016). Eventually, cognitive skills (social intelligence, flexibility, creativity) are instead becoming a necessary albeit not sufficient condition to find good and well-paid occupations (Deming, 2017). Especially in times of a globalization of economic activities and of an automation of production, soft skills may be crucial for their impact on productivity, economic growth and career perspectives. We argue that a truly interdisciplinary approach is needed to better understand the role of soft and cognitive skills (Bakhshi et al., 2017). For example, how can we measure creativity and teamwork? How do cognitive and soft skills relate to STEM and Humanities? Can we complement expert consensus forecasts with the most recent machine learning techniques to predict future job trends across the different demands for skills? Which policies can we derive for R&D, education, productivity and employment?
Keywords
technological change, soft skills, labor markets
Reference faculty: Massimo Riccaboni, Francesco Serti
References
  • Bakhshi, H., Downing, J., Osborne, M., & Schneider, P. (2017). The future of skills: employment in 2030. London: Nesta, Oxford-Martin, Pearson.
  • Beaudry, Paul, David A. Green, and Benjamin M. Sand (2016). “The Great Reversal in the Demand for Skill and Cognitive Tasks”, Journal of Labor Economics, 34, 199–247.
  • Deming, D. (2017). “The growing importance of social skills in the labor market”, The Quarterly Journal of Economics, 132, 1593-1640.
Technology diffusion, productivity of firms, and endogenous economic growth
Abstract
Recent technological developments (e.g. artificial intelligence, 3D printing, blockchain) are already reshaping the organization of the global economy, while cross-country and within-country differences in productivities and incomes persist. In fact, the adoption of Information and Communication Technologies (ICT) in the latest 15 years seems to coincide with a slowdown in aggregate productivity, especially in more advanced economies (Comin and Mestieri, 2018). This is apparently in contradiction with expectations that investment in innovation and productive knowledge shall foster general welfare and economic growth (Romer, 1990). Several explanations have been proposed leading to different conclusions (Haldane, 2017), ranging from a more pessimistic scenario of a “Great Stagnation” to an optimistic quest for errors of measurement in national accounts. Interestingly, the role of the organization of firms and the competitive environment in which they operate, within and across countries, has been so far relatively neglected. How could firms and markets foster or hinder the diffusion of technology? How could a different organization of firms and markets guarantee that technology improvements keep contributing to “perpetually rising standards of living” (Grossman and Helpman, 1994)?
Keywords
endogenous growth theory, firm-level, technology diffusion
Reference faculty: Massimo Riccaboni, Armando Rungi
References
  • Andrews, D. Criscuolo C. and Gal P. (2016), “The Best versus the Rest: The Global Productivity Slowdown, Divergence across Firms and the Role of Public Policy”, OECD Productivity Working Papers, No. 5
  • Comin D. and Mestieri M. (2018). “If Technology Has Arrived Everywhere, Why Has Income Diverged?” American Economic Journal: Macroeconomics, 10 (3): 137-78.
  • Grossman G. M. and Helpman E. (1994). “Endogenous Innovation in the Theory of Growth,” Journal of Economic Perspectives, American Economic Association, vol. 8(1), pages 23-44.
  • Haldane A. G. (2017). Productivity Puzzles. Speech held by the Chief Economist of the Bank of England at the London School of Economics on 20 March 2017, available online at https://www.bankofengland.co.uk/speech/2017/productivity-puzzles.
  • Romer, P. M. (1990). "Endogenous Technological Change," Journal of Political Economy, 98:5, S71–102.
Production networks and firms’ outcomes
Abstract
The recent research on firm-to-firm transactions has shown the importance of the structure of a production network for explaining the outcomes of the firms that participate to its formation (Bernard and Moxnes, 2018). For example, more and better suppliers may affect the performance of downstream firms both in domestic and international production networks (Bernard et al., 2016). Yet information frictions may play a role in how links among firms establish and how firms interact, especially when buyers and sellers are distant from each other and when their products are innovative and complex, in which case reciprocal knowledge and trusted partners seems to facilitate the establishment of more productive trade relationships (Chaney, 2016). Eventually, the choices of participating firms collectively determine an input-output macroeconomic equilibrium that could be based on large differences in size and productivity across an economy (Oberfield, 2018). In a context of interdependence between economic agents, there is a large number of unanswered research questions that we can ask ourselves. To what extent and how a firm’s productivity is affected by variations in the performance of its (direct and indirect) suppliers and customers? What are the determinants of the formation of the trade links between firms and, therefore, of inputs adoption? Do firms use existing links to learn about new productivity-enhancing links? Is network closure/clustering between firms important in stimulating cooperation along the supply chain and, therefore, in fostering the adoption and the efficient exploitation of innovative production techniques? What is the role of multinational enterprises in production networks?
Keywords
production networks, network formation, firm-level outcomes
Reference faculty: Massimo Riccaboni, Armando Rungi, Francesco Serti
References
  • Bernard, A. and Moxnes, A. (2018). Networks and trade. Annual Review of Economics, vol. 10:65-85
  • Bernard, A., A. Moxnes and Yukiko U. Saito, (2016). Production Networks, Geography and Firm Performance. CEP Discussion Papers dp1435, Centre for Economic Performance, LSE.
  • Chaney, T. (2016). Networks in international trade. In Y. Bramoulle, A. Galeotti and B. Rogers (eds.), Oxford Handbook of the Economics of Networks, Oxford: Oxford University Press.
  • Oberfield, E. (2018). A theory of input–output architecture. Econometrica, 86 (2), 559–589.
Machine learning and econometrics
Abstract
A recent trend of research aims to combine methods from machine learning and econometrics for the analysis of economic and business data (Varian, 2014). Typically, machine learning exploits training data and prior knowledge to teach a model how to predict - in the best possible way, given the information available - an output variable as a function of an input vector (Vapnik, 1998). However, classical machine learning methods do not address other issues such as endogeneity and causal inference (Imbens and Rubin, 2015), which have been the traditional field of study of econometrics. In fact, causal inference is particularly important for the analysis of micro/macroeconomic data, and eventually for policy evaluation (Athey and Imbens, 2017). According to (Varian, 2014), machine learning can learn from econometrics how: to deal with data that are not independent and identically distributed; to perform causal inference; to work with instrumental variables. Similarly, econometrics can learn from machine learning how: to train/validate/test and avoid overfitting; to apply non-linear estimation methods such as neural networks; to perform variable selection; to work with extremely large datasets. An example of the successful interaction between the two disciplines is provided by the recent causal tree algorithm (Athey and Imbens, 2016) for the estimation of heterogeneous causal effects. Several other algorithms are expected to be developed in the future.
Keywords
machine learning, econometrics, causal inference
Reference faculty: Massimo Riccaboni, Giorgio Gnecco, Armando Rungi
References
  • Athey, S., Imbens, G. W. (2016). Recursive partitioning for heterogeneous causal effects. Proceedings of the National Academy of Sciences 113:7353-7360.
  • Athey, S., Imbens, G. W. (2017). The state of applied econometrics: Causality and policy evaluation. Journal of Economic Perspectives 31, 3-32.
  • Imbens, G.W., and Rubin, D.B. (2015). Causal Inference for Statistics, Social, and Biomedical Sciences. An Introduction. Cambridge University Press.
  • Vapnik, V. N. (1998). Statistical Learning Theory. Wiley.
  • Varian, H.R. (2014). Big data: New tricks for econometrics. Journal of Economic Perspectives 28, 3-28.
Financial shocks, firms’ financial constraints, and production networks
Abstract
Modern economies are organized as webs of specialized producers engaged in exchanges of tangible and intangible inputs. Hence, a network perspective can provide insights on how local shocks can propagate heterogeneously across the entire economy, therefore generating aggregate and granular fluctuations, whose magnitude and volatility may in turn depend on the topology of the underlying production network (Carvalho, 2014; Acemoglu et al., 2016). This is particularly true for the nature of the propagation of financial shocks throughout the real economy. An exogenous financial shock affecting a single firm or a category thereof, e.g. at the beginning of a financial crisis, can reverberate through a supply chain (Escaith et al., 2011; Altomonte et al., 2012) increasing the initial magnitude of the shock and the volatility of the adjustments. More in general, financial constraints at the firm level can transmit heterogeneously across national borders and affect international business cycles (di Giovanni et al., 2018). How can we estimate the direct and indirect impact of a future shock on the real economy? How much responsive are international business cycles to firms’ financial constraints? Which policies could be enhanced for limiting the damage on the general welfare and alleviate the firms’ financial constraints?
Keywords
financial constraints, production networks, global value chains
Reference faculty
Massimo Riccaboni, Armando Rungi
References
  • Acemoglu D., Carvalho V. M., Ozdaglar A., Tahbaz‐Salehi A. (2016). “The Network Origins of Aggregate Fluctuations”, Econometrica, vol. 80(5): 1977-2016.
  • Altomonte C., Di Mauro F., Ottaviano G., Rungi A and Vicard V. (2013) “Global value chains during the Great Trade collapse: a bullwhip effect” in: Firms in the international economy: firm heterogeneity meets international business. CESifo Seminar Series. MIT Press, pp. 277-308
  • Carvalho, V. M. (2014). “From Micro to Macro via Production Networks.” Journal of Economic Perspectives, 28 (4): 23-48.
  • di Giovanni J, Levchenko A. J., and Mejean I. (2018). "The Micro Origins of International Business-Cycle Comovement," American Economic Review, vol 108(1), pages 82-108.
  • Escaith H., Lindenberg N. and Miroudot S. (2011). “Global Supply Chains, the Great Trade Collapse and Beyond: More Elasticity or More Volatility?” in “Recovery and Beyond: Lessons for Trade Adjustment and Competitiveness" Editors: F. di Mauro et B. Mandel. European Central Bank.
Interacting Reinforced Stochastic Processes (RSPs) and their applications
Abstract
Many scientific fields are interested in the stochastic evolution of networks of interacting agents that develop asymptotically a common behavior, a phenomenon typically denoted as “synchronization”. Examples of this type of systems can be found in a large number of scientific areas, such as economics, neuroscience, social and computer science. In all these research areas, the main goals are: (1) to figure out when a synchronization may emerge and (2) to investigate the interplay between the network topology and the collective dynamics followed by the agents. Recently these two goals have been accomplished for networks of agents whose behavior can be modeled by Stochastic Processes that evolve following a Reinforcement mechanism (RSPs), which is when the probability of occurrence of a given event increases with the number of its occurrences in the past. A well-known example of RSP is the Pòlya urn. From a theoretical point of view, this project aims at studying other dynamics for the agents' behaviors and other mechanisms of interaction among them, and, from an applicative point of view, aims at employing the obtained results in order to analyze real interacting systems. This project requires a candidate with strong skills in probability theory and mathematical analysis and statistics.
Keywords
Interacting Random Systems; Reinforced Stochastic Processes; Urn Models; Complex Networks; Preferential Attachment; Weighted Empirical Means; Synchronization; Asymptotic Normality.
Reference Faculty 
Irene Crimaldi (AXES)
References
  • G. Aletti - I. Crimaldi - A. Ghiglietti, Networks of reinforced stochastic processes: asymptotics for the empirical means, forthcoming in Bernoulli.
  • I. Crimaldi - P. Dai Pra - P-Y. Louis - I. G. Minelli (2019), Synchronization and functional central limit theorems for interacting reinforced random walks, Stochastic Processes and their Applications, 129(1), 70-101.
  • G. Aletti - I. Crimaldi - A. Ghiglietti (2017), Synchronization of reinforced stochastic processes with a network-based interaction, The Annals of Applied Probability, 27(6), 3787-3844.
Management Science and behavioral strategy: a view on debiasing and decision making process

Abstract
Decision making process is a main part of managing activities across all kinds of firms. Every decision is apt to product consequences for the firm, also impacting its performances. In the last decades, management research is increasingly focusing on the behavioral approach: integrating psychological theories and methods to management science can increase the understanding of how cognitive and emotional processes work and shape individuals’ decisions and actions. This line of research shall study how cognitive and social psychology they apply to strategic management theory and practice (namely, behavioral strategy). The behavioral approach uses the cognitive psychology, which is a branch of psychology that seeks to understand the internal mental processes of thought. The main research theme is business organization, not only the man. The candidate will embrace topics in the existing core of behavioral strategy (such as decision biases and cognitive schema), while encouraging innovations and de-biasing actions in a managerial context.
Reference Faculty
Nicola Lattanzi, Emiliano Ricciardi
References

  • Powell, T. C., Lovallo, D., & Fox, C. R. (2011). Behavioral strategy. Strategic Management Journal, 32(13), 1369-1386.
  • Lattanzi, N. (2013). Management Science and Neuroscience Impact. Decision Making Process, Entrepreneurship and Business Strategy. McGraw-Hill.
  • Ricciardi, E., Bonino, D., Gentili, C., Sani, L., Pietrini, P., & Vecchi, T. (2006). Neural correlates of spatial working memory in humans: a functional magnetic resonance imaging study comparing visual and tactile processes. Neuroscience, 139(1), 339-349.
  • Intesa San Paolo Innovation Center & IMT School for Advanced Studies Lucca (2018). Innovation Trend Report: Neuroscience Impact. Brain and Business.
Business performance and soft skills

Abstract
Human beings manage all organizations; they are made of men and are imperfect systems. Firms’ long-term success greatly depends on how managers can select and govern human resources, also being able to understand the nature and complexities of human beings as multi-faceted individuals. Individuals also influence firm performances: employing the right type of person (with certain skills, behaviors and abilities) can help an organization increase productivity and maintain a competitive advantage. This line of research studies how personality influences job and firm performances.
Reference Faculty
Nicola Lattanzi, Massimo Riccaboni. Andrea Morescalchi
References

  • Barrick, M. R., Stewart, G. L., & Piotrowski, M. (2002). Personality and job performance: test of the mediating effects of motivation among sales representatives. Journal of Applied Psychology, 87(1), 43.
  • Seriki, O. K., Nath, P., Ingene, C. A., & Evans, K. R. (2018). How complexity impacts salesperson counterproductive behavior: The mediating role of moral disengagement. Journal of Business Research.
  • Lattanzi, N., Menicagli, D., & Dal Maso, L. (2016). Neuroscience Evidence for Economic Humanism in Management Science: Organizational Implications and Strategy. Archives italiennes de biologie, 154(1), 25-36.
SMEs, family businesses, and industrial districts

Abstract
Family firms are a key component of the European economy, both for their high number and contributes to GDP and occupation. Although family businesses are not an Italian peculiarity, they strongly characterize the Italian economy. Research has not currently reached a unique paradigm on the relationships between the involvement of the family in the ownership and management of a firm. This line of research investigates the relationship between family firms and industrial districts, which is currently underdeveloped in the academic literature. The candidate will work both on developing the theoretical models and will analyse data using quantitative (econometrics and social network analysis) and qualitative methods (case studies).
Reference Faculty
Nicola Lattanzi, Marco Paggi e Alberto Bemporad, Armando Rungi
References

  • Cucculelli, M., Storai, D. (2015). Family Firms and Industrial Districts: Evidence from the Italian Manufacturing Industry. Journal of Family Business Strategy, 6(4), p. 234-246
  • Lattanzi, N. (2017). Le aziende familiari: Generazioni Società Mercato. G Giappichelli Editore.
Business models for emerging markets: a network theory contribution

Abstract
Fintech (Financial Technology) includes a wide set of technologies and innovations that are revolutionizing traditional financial services. Blockchain's uses have recently evolved into many applications, such as banking, financial markets, insurance and leasing contracts. Today, blockchain has potential for application in various business fields, including accounting and in certifying financial statements. This line of research will follow blockchain’s fundamental concepts, providing perspectives on its challenges and opportunities in business, management and accounting practices, also using complex systems for management science.
Reference Faculty
Nicola Lattanzi, Diego Garlaschelli
References

  • De Bruijn, H., & Ten Heuvelhof, E. (2018). Management in networks. Routledge.
  • Dai, J., & Vasarhelyi, M. A. (2017). Toward blockchain-based accounting and assurance. Journal of Information Systems, 31(3), 5-21.
  • Fanning, K., & Centers, D. P. (2016). Blockchain and its coming impact on financial services. Journal of Corporate Accounting & Finance, 27(5), 53-57.