Alessandro Morari, Ph.D.
Manager and Senior Research Scientist
AI for Code
IBM Research
Highly accomplished manager and researcher with a proven track record of leading teams and projects in Applied Machine Learning and High Performance Computing.
In 2019 I spearheaded the AI for Code initiative for IBM Research, building a talented team of researchers and leading multiple research projects in the field. AI for Code produced top-tier publications, patents, and considerably advanced IBM know-how, making it a leader in the space.
In the High Performance Computing space, I led the team responsible for the Operating System performance and scalability of the Summit and Sierra supercomputers, the world's fastest supercomputers in 2018, part of a $300M+ procurement for the Department of Energy.Â
I also developed and taught the first High Performance Machine Learning graduate course for the NYU Courant Institute of Mathematical Science, teaching graduate students how to apply computer architecture and high performance computing to scale Machine Learning algorithms. This curriculum is now being regularly taught at NYU and multiple other universities.
While at Pacific Northwest National Laboratory, I was one of the three researchers to develop GEMS, a high performance graph database that resulted in multiple top-tier publications and the spin-off of Trovares, a startup offering highly scalable graph analytics on the cloud. For this work I won the PNNL Outstanding performance Award.
I regularly advise startups, entrepreneurs, and small companies regarding technological trends and digital strategy.
Education
Ph.D. in Computer Architecture, Polytechnic University of Catalunya.
M.Sc. in Computer Engineering, University of Rome Tor Vergata.
B.Sc. in Computer Engineering, Roma Tre University.
Current Research
Towards Learning (Dis)-Similarity of Source Code from Program Contrasts. Authors: Yangruibo Ding, Luca Buratti, Saurabh Pujar, Alessandro Morari, Baishakhi Ray, Saikat Chakrabort. ACL 2022.
VELVET: a noVel Ensemble Learning approach to automatically locate VulnErable sTatements. Authors: Yangruibo Ding, Sahil Suneja, Yunhui Zheng, Jim Laredo, Alessandro Morari, Gail Kaiser, Baishakhi Ray. SANER 2022.
D2a: A dataset built for ai-based vulnerability detection methods using differential analysis. Authors: Yunhui Zheng, Saurabh Pujar, Burn Lewis, Luca Buratti, Edward Epstein, Bo Yang, Jim Laredo, Alessandro Morari, Zhong Su. ICSE-SEIP 2021.
Data-Driven AI Model Signal-Awareness Enhancement and Introspection. Authors: Sahil Suneja, Yufan Zhuang, Yunhui Zheng, Jim Laredo, Alessandro Morari. arXiv preprint arXiv:2111.05827.
Contrastive Learning for Source Code with Structural and Functional Properties. Authors: Yangruibo Ding, Luca Buratti, Saurabh Pujar, Alessandro Morari, Baishakhi Ray, Saikat Chakraborty. arXiv preprint arXiv:2110.03868.
Software Vulnerability Detection via Deep Learning over Disaggregated Code Graph Representation. Authors: Yufan Zhuang, Sahil Suneja, Veronika Thost, Giacomo Domeniconi, Alessandro Morari, Jim Laredo. arXiv preprint arXiv:2110.03868.
Probing Model Signal-Awareness via Prediction-Preserving Input Minimization. Authors: Sahil Suneja, Yunhui Zheng, Yufan Zhuang, Jim Laredo, Alessandro Morari. ESEC/FSE 2021.
Towards Reliable AI for Source Code Understanding. Authors: Sahil Suneja, Yunhui Zheng, Yufan Zhuang, Jim A Laredo, Alessandro Morari. ACM Symposium on Cloud Computing 2021. SoCC 2021.
Cush: Cognitive scheduler for heterogeneous high performance computing system. Authors: Giacomo Domeniconi, Eun Kyung Lee, V Venkataswamy, S Dola. DRL4KDD 2019.