Hava Siegelmann Women's Chapter Chair University of Massachusetts - Amherst, USA

Hava Siegelmann, is a Provost Professor of the University of Massachusetts, Amherst, as well as a tenured full professor at the College of Information and Computer Sciences (CICS), a core faculty of Brain and Cognitive Sciences, and the director of the BINDS Lab. Her research focuses on advancing the state of neural networks, and on the emerging field of Lifelong Learning, which is at the cutting edge of Machine Learning and Artificial Intelligence. In 2016, Siegelmann took a leave from UMass to develop and direct DARPA’s extensive, groundbreaking Lifelong Learning Machines (L2M) program. The field of Lifelong Learning encompasses research, development, and implementation of robust computational systems capable of learning during field time and applying previous experience to navigate novel circumstances without reprogramming or retraining. In addition, L2M systems can adapt to component upgrades without retraining ab initio - based on prior experience and learning capabilities. Siegelmann’s second major DARPA program is Guaranteeing AI Robustness to Deception (GARD), which works on evaluating, developing defenses against new generation attacks against AI, and assuring system robustness and resilience. She initiated a few other programs, including in AI safety, distributed collaboration, and biomedical engineering.

Siegelmann’s prior research into neural processes and architecture has led to superior, increasingly adaptive algorithms, models, and systems. Her work in modeling geometric neural clusters resulted in the Support Vector Clustering algorithm (with Vladimir Vapnik) - one of the most widely used clustering algorithms in industry. Siegelmann developed the first, and one of the few extant compilers capable of constructing an equivalent recurrent neural network from any program. Her efficient symbolic driven neural architectures have been utilized in complex industrial settings such as radar installations and nuclear power plants since the 1990’s – some of the earliest, most complex, and most reliable neural networks to date.

Siegelmann is the inventor of Super-Turing theory, a subfield in AI establishing an alternate form of computation from that of the more familiar Turing type. Her work provides a foundational understanding of biological computation and their learning systems which combine learning and computation, just like these she later introduced in her DARPA L2M program. In 2015, the NIH named her one of 16 presidential BRAIN Initiative awardees for her work on energy constrained brain activation. In 2016, the International Neural Network Society named Siegelmann their Donald O. Hebb Awardee for lifelong accomplishments in the field of neural network learning; that same year, the Institute of Electrical and Electronics Engineers named her an IEEE distinguished lecturer. Siegelmann was made an IEEE fellow in 2017. In 2020, The department of Defense bestowed upon her one of the highest honors for citizens – the Meritorious Public Service Medal. In 2022, the University of Massachusetts named her Provost Professor. At the same year, the INNS welcomed her to join as a Fellow.

Much of Siegelmann’s work in AI is biologically inspired. In addition to her work in artificial systems learning, Siegelmann’s 2010-15 memory reconsolidation models were the only ones capable of using experience for handling dynamic environments; this work is now also a core element of numerous L2M efforts. Her biomedical contributions include shedding light on aspects of human brain architecture leading to abstract thought, the first full elucidation of the suprachiasmatic nucleus (biological clock), computational models of addiction, and advanced human-machine interfaces that significantly enhance multitasking performance. Her engineering background has allowed her to take part in analog hardware design in emerging lifelong learning computational systems and develop a new DARPA program to produce lightweight, portable sensors capable of quickly identify toxic opioids at a distance.

As a professor, Siegelmann has taught a variety of highly innovative, interdisciplinary classes that incorporate information systems, mathematics, biology, and neuroscience with direct application to Lifelong Learning. Siegelmann has been serving as the Co-Chair of the University’s Diversity Council, focusing on supporting young researchers, minorities, and women to advance in academia, and is the founder and chair of the Women’s chapter of the International Neural Network Society. She also has years of experience consulting with industry, creating educational programs domestically and internationally, in academic fundraising, and in educational and research administration.