Alessandro Oltramari, Ph.D.
Hybrid intelligent systems: combining knowledge representation and machine learning
“Knowledge representation can drive learning algorithms, enable high-level reasoning and generalization mechanisms, make real-time data analysis more robust.”
I work on knowledge-based intelligent systems. My primary interest is to investigate how semantic resources like computational ontologies and lexical databases can be used to augment learning algorithms, and help machines to make sense of the world. I strive to make progress in explainable AI, which I’m convinced can be achieved only by designing systems that integrate human-accessible machine representations with neural networks.
- Research Associate, Carnegie Mellon University (Pittsburgh, USA)
- Visiting Research Associate, Princeton University (Princeton, USA)
- Research Fellow, Laboratory for Applied Ontology (CNR, Trient, Italy)
A. Oltramari et al. (2021)Knowledge-driven Data Construction for Zero-shot Evaluation in Commonsense Question Answering
- K. Ma, F. Ilievsky, J. Francis, Y. Bisk, E. Nyberg, A. Oltramari
- Proceeding of The Thirty-Fifth AAAI Conference on Artificial Intelligence (AAAI-21)
- February 2-9 2021
A. Oltramari et al. (2020)Neuro-Symbolic Architectures for Context Understanding
- Oltramari, A., Francis J, Henson C, Ma K., Wickramarachchi R.
- In “Knowledge Graphs for eXplainable Artificial Intelligence: Foundations, Applications and challenges”
- Studies on the Semantic Web (47), Ed. I. Tiddi, F. Lecué, P. Hitzler.
A. Oltramari et al. (2020)Cognitive Twin: A Cognitive Approach to Personalized Assistants
- S. Somers, A. Oltramari, C. Lebiere
- Proceedings of AAAI Spring Symposia 2020 (AAAI-MAKE: Combining Machine Learning and Knowledge Engineering in Practice)
- March 23-25 2020
A. Oltramari (2019)Artificial Intelligence Within the Bounds of Ontological Reason
- In “Ontology Makes Sense”
- Vol. 316 of Frontiers in Artificial Intelligence and Applications, (pp. 37-48)
- IOS Press
A. Oltramari et al. (2018)A semantic framework for the analysis of privacy policies
- A. Oltramari, D. Piraviperumal, F. Schaub, S. Wilson, S. Cherivirala, T. Norton, & N. Sadeh
- PrivOnto: A semantic framework for the analysis of privacy policies
- Semantic Web 9 (2), p. 185–203
A. Oltramari et al. (2016)Security Taxonomies of Industrial Control Systems
- S. A. Flowers, Sidney C. Smith, A. Oltramari
- E. Colbert, A. Kott (eds.)
- Cyber-security of SCADA and Other Industrial Control Systems
N.B. Asher et al. (2015)Ontology-based Adaptive Systems of Cyber Defense
- N.B. Asher, A. Oltramari, C. Gonzalez, R. Erbacher
- STIDS 2015 (10th International Conference on “Semantic Technology for Intelligence, Defense, and Security”, Fairfax (VA), 18-20 November, CEUR Workshop Proceedings
C. Gonzalez et al. (2014)Cognition and Technology
- C. Gonzalez, C., N. Ben-Asher, A. Oltramari, C. Lebiere
- In Wang, C., Kott, A., Erbacher, R. (Eds.), Cyber Defense and Situational Awareness, Springer-Verlag
A. Oltramari et al. (2013)New Trends of Research in Ontologies and Lexical Resources
- A. Oltramari, P. Vossen, L. Qin, E. Hovy
A. Oltramari et al. (2013)Senso Comune: A Collaborative Knowledge Resource for Italian
- A. Oltramari, G. Vetere, E. Jezek, I. Chiari, F. Zanzotto, M. Nissim, A.
- The People's Web Meets NLP: Collaboratively Constructed Language Resources, p. 45-67
A. Oltramari & C. Lebiere (2012)Pursuing Artificial General Intelligence By Leveraging the Knowledge Capabilities Of ACT-R
- In AGI 2012 (5th International Conference on “Artificial General Intelligence”), Oxford (UK), 199-208.
A. Oltramari (2011)An Introduction to Hybrid Semantics: The Role of Cognition in Semantic Resources. Modeling, Learning, and Processing of Text Technological Data Structures Studies
- Computational Intelligence. Volume 370, p. 97-109
C. R. Huang et al. (2010)Ontology and the Lexicon
- C. R. Huang, N. Calzolari, A. Gangemi, A. Lenci, A. Oltramari, L. Prévot (Editors)
- Cambridge University Press
A. Oltramari et al. (2002)Restructuring WordNet's Top-Level: The OntoClean approach
- A. Oltramari, A. Gangemi, N. Guarino, C. Masolo / R. S. Araujo (editor)
- Proc. of LREC 2002 (The 4th International conference on Language Resources and Evaluation – Workshop OntoLex 2002 – Ontologies and Lexical Resources, 27 May, Las Palmas de Gran Canaria, Spain), p. 17-22.
A. Oltramari et al. (2002)Sweetening Ontologies with DOLCE
- A. Gangemi, N. Guarino, C. Masolo, A. Oltramari, and Schneider, L.
- 2002, October. Sweetening ontologies with DOLCE. In International Conference on Knowledge Engineering and Knowledge Management (pp. 166-181). Springer, Berlin, Heidelberg.
Interview with Alessandro Oltramari, Ph.D.
Senior Research Scientist in the areas of Applied Ontology, Knowledge Representation and Reasoning, Cognitive Systems
Please tell us what fascinates you most about research.
For me research is a journey of the mind: it’s an exploration of a problem space, by means of which a map of solutions can be drawn. As any human endeavor, research is accompanied by a full variety of emotions: Joy, frustration, disappointment, surprise are part of any scientist’s life. I should add that research for me has also been a journey in the literal sense of the term: I left Italy in search of better opportunities to fulfill my career goals. In Bosch Research I found my ideal “research home.”
What makes research done at Bosch so special?
There are several aspects that make working at Bosch special, from the multi-cultural environment to the open dialogue with upper management. But the essential feature for me is the balance between business-driven research problems and scientific investigations born out of research teams. Striking a balance between short-term and long-term research is facilitated in Bosch by a rigorous and yet flexible practice that promotes clear goals setting, business model construction and validation, collaborative planning and task execution, retrospective review, and open discussion.
What research topics are you currently working on at Bosch?
I’m investigating on how we can build more robust intelligent assistants by integrating heterogeneous knowledge sources (electronic documents, information on the Internet, social networks) with machine learning algorithms. In particular, I’m looking at this topic from a concrete application angle in three different domains: context-aware chatbots for customer services, decision support systems for (semi-)autonomous cars, and scene understanding in smart environments.
What are the biggest scientific challenges in your field of research?
Explainable artificial intelligence is the number one challenge. In order to progress towards explainable AI, it is necessary to design hybrid systems that integrate human-accessible machine representations with neural machines. Currently, a general framework of integration between deep networks and knowledge graphs doesn’t really exist.
How do the results of your research become part of solutions "Invented for life"?
Enabling explainable AI means endowing intelligent machines with semantic transparency, in terms of both internal functioning and correlation between input and output. As humans learn how to trust each other by sharing knowledge, explainable systems will make human machine interaction more trustworthy and personalized, hence suitable to improve human life in many areas where AI is crucial (healthcare, mobility, etc.).