Seminar IDIA/FCS

When designing complex systems, we need to consider multiple trade-offs at various abstraction levels and scales, and choices of single components need to be studied jointly. For instance, the design of future mobility solutions (e.g., autonomous vehicles, micromobility) and the design of the mobility systems they enable are closely coupled. Indeed, knowledge about the intended service of novel mobility solutions would impact their design and deployment process, while insights about their technological development could significantly affect transportation management policies. Optimally co-designing sociotechnical systems is a complex task for at least two reasons. On one hand, the co-design of interconnected systems (e.g., large networks of cyber-physical systems) involves the simultaneous choice of components arising from heterogeneous natures (e.g., hardware vs. software parts) and fields, while satisfying systemic constraints and accounting for multiple objectives. On the other hand, components are connected via collaborative and conflicting interactions between different stakeholders (e.g., within an intermodal mobility system). In this talk, I will present a framework to co-design complex systems, leveraging a monotone theory of co-design and tools from game theory. The framework will be instantiated in the task of designing future mobility systems, all the way from the policies that a city can design, to the autonomy of vehicles as part of an autonomous mobility-on-demand service. Through various case studies, I will show how the proposed approaches allow one to efficiently answer heterogeneous questions, unifying different modeling techniques and promoting interdisciplinarity, modularity, and compositionality. I will then discuss open challenges for compositional systems design optimization, and present my agenda to tackle them. bio: | Gioele is the Rudge (1948) and Nancy Allen Assistant Professor at Massachusetts Institute of Technology. He is a PI in the Laboratory for Information and Decision Systems (LIDS), the Department of Civil and Environmental Engineering (CEE), and an affiliate faculty with the Institute for Data, Systems and Society (IDSS). He received his doctoral degree in 2024 from ETH Zurich and holds both a BSc. and an MSc. in Mechanical Engineering and Robotics, Systems and Control from ETH Zurich. Before joining MIT as a faculty, he was a Postdoctoral Scholar at Stanford University (January to June 2024), and held various visiting positions at nuTonomy Singapore (then Aptiv, now Motional), Stanford, and MIT. Driven by societal challenges, the goal of his research is to develop efficient computational tools and algorithmic approaches to formulate and solve complex, interconnected system design and autonomous decision-making problems. His interests include the co-design of complex systems, all the way from future mobility systems to autonomous systems, compositionality in engineering, planning and control, and game theory. He is the creator of Autonomy Talks (an International seminar series promoting a diverse research exchange on autonomy), as well as a lead organizer for the seminal workshops Compositional Robotics: Mathematics and Tools, and Co-Design and Coordination of Future Mobility Systems at IEEE ICRA and ITSC, respectively. He is the recipient of the ETH Medal for his doctoral thesis, of a paper award at the 4th Applied Category Theory Conference, the Best Paper Award (1st Place) at the 24th IEEE International Conference on Intelligent Transportation Systems (ITSC), DoE, MIT, Sidara, and Amazon awards.

A broad class of models deal with spaces of directed paths avoiding some dynamic obstacles. (Examples include concurrency, in computer science, and collision avoidance in control theory.) The topology (say, the homotopy type) of these spaces is not yet well understood. It turns out, the study of these spaces is in many respects parallel to the theory of linear subspace arrangements. I will outline some recent results, and the key tools used to obtain them, - in particular, the apparatus of diagrams of spaces. video: https://inria.webex.com/inria/j.php?MTID=m5eae99d506bf763fbc41b4f5a1a1d61d

Blockchains have served as decentralized platforms to transform the way we think about databases, distributed systems and finance in a world where people can interact without knowing each other and across borders without government intermediaries. Nevertheless, an historic perspective of the ideas that brought us here can be traced back to the birth of decentralized systems in the 1970s with roots coming from research in cryptography, distributed systems and databases. A fascinating history where new notions were invented along the way, requiring the design of new algorithms, and the understanding of scientific principles about what can and cannot be done by a collection of computers communicating with each other.

Formal verification of deep learning models remains challenging, and yet they are becoming integral in many safety-critical autonomous machines. We present a framework for certifying the end-to-end safety of learning-enabled autonomous systems using perception contracts. The method flows from the observation that learning-based perception is often used for state estimation, and the error characteristics of such estimators can be succinct, testable, and amenable to effective formal analysis. Our framework constructs approximations of perception models, using system-level safety requirements and program analysis. Mathematically proving that a given perception model conforms to a contract will be challenging, but empirical conformance measures can provide confidence levels to safety claims. We will discuss recent applications of this framework in creating low-dimensional, intelligible contracts and end-to-end safety certificates for vision-based lane-keeping and auto-landing systems, as well as a number for future research directions.

Autonomous systems are emerging as a driving technology for countlessly many applications. Numerous disciplines tackle the challenges toward making these systems trustworthy, adaptable, user-friendly, and economical. On the other hand, the existing disciplinary boundaries delay and possibly even obstruct progress. I argue that the nonconventional problems that arise in designing and verifying autonomous systems require hybrid solutions at the intersection of learning, formal methods, and controls. I will present examples of such hybrid solutions in the context of learning in sequential decision-making processes. These results offer novel means for effectively integrating physics-based, contextual, or structural prior knowledge into data-driven learning algorithms. They improve data efficiency by several orders of magnitude and generalizability to environments and tasks that the system had not experienced previously. I will conclude with remarks on a few promising future research directions