Jorge Lobo

Jorge Lobo

Universitat Pompeu Fabra

Engineering Sciences

I am an ICREA Research Professor in the Department of Information and Communication Technologies at UPF, and a Visiting Professor at Imperial College (London) Dept. of Computing. After receiving a PhD in Computer Science from University of Maryland (1990), I took a faculty position at the University of Illinois in Chicago, where I obtained tenure (1997). In 1998 I was invited to join Bell Labs’ newly-formed Network Computing group, to help them integrate my work on rule system as logic program (then purely theoretic) into applications in the burgeoning field of soft-switches. The fruitful endeavor (Lucent NetMon, 2000) served to expand my interest in basic research to its real-world applications, and in 2001 I became Principal Architect at Teltier Technologies, a start-up company founded by researchers from Bell Labs (now part of Cisco Systems). I returned to full time research as Staff Member at IBM TJ Watson lab (2004-14), and in 2009 became an ACM Distinguished Scientist.

Research interests

My early work in formal logic has been the basis of all my research, which is at the interface of theory and practice in computer science. Broadly, my goal is to automate the complex task of managing distributed systems (eg, networks of computers) in ways that have a solid theoretical foundation (rather than by ad-hoc means). Central to this effort is the concept of policies: collections of rules controlling the system’s behavior (eg, who can access private data). A major challenge is that amassing rules leads to conflicts (in access control, scheduling tasks, etc). My work has shown that the semantics of policies, as well as potential conflicts, can be defined axiomatically using disjunctive logic programs. Beyond conflict detection, the logic programming formulation provides a computational framework for conflict resolution, policy analysis and verification. Recent work focuses on how rules can be learned from examples (ie, inferred) using inductive logic programming techniques.

Selected publications

- Cunnington, D., Law, M., Lobo, J. et al. FFNSL: Feed-Forward Neural-Symbolic Learner. Mach Learn 112, 515–569 (2023).
- Abu Jabal A, Bertino E, Lobo J, et al. 2023, 'FLAP - A Federated Learning Framework for Attribute-based Access Control Policies', in Proceedings of the Thirteenth ACM Conference on Data and Application Security and Privacy (CODASPY '23). Association for Computing Machinery, New York, NY, USA, 263–272.
- Cunnington D, Law M Lobo J & Russo A 2023, 'Neuro-symbolic learning of answer set programs from raw data', in Proceedings of the Thirty-Second International Joint Conference on Artificial Intelligence, 3586-3596.

Selected research activities

Steering Committee Chair, The ACM Symposium Series on Access Control Models and Technologies (SACMAT).
Visiting Professor, Imperial College London (Honorary).