Hava Siegelmann

Updated on Oct 04, 2024

Edit

Comment

Alma mater Rutgers University Name Hava Siegelmann		Role Computer scientist

Institutions University of Massachusetts Amherst Thesis Foundations of Recurrent Neural Networks (1993) Education Technion – Israel Institute of Technology Books Neural networks and analog computation Fields Computer Science, Neuroscience, Systems biology, Biomedical engineering
Doctoral advisor Eduardo D. Sontag

Lifelong learning machines (L2M) - Hava Siegelmann keynote at HLAI

Hava Siegelmann is a professor of computer science, and a world leader in the fields of Artificial Intelligence, Machine Learning, and Computational Neuroscience. Her academic position is in the school of Computer Science and the Program of Neuroscience and Behavior at the University of Massachusetts Amherst; she is the director of the school's Biologically Inspired Neural and Dynamical Systems Lab.

Artificial Intelligence Colloquium: Lifelong and Robust Machine Learning

Biography

Siegelmann is an American computer scientist who founded the field of super-Turing computation. For her lifetime contribution to the field of Neural Networks she is the recipient of the 2016 Donald Hebb Award. She earned her PhD at Rutgers University, New Jersey, in 1993.

In the early 1990s, she and Eduardo D. Sontag proposed a new computational model, the Artificial Recurrent Neural Network (ARNN), which has been of both practical and mathematical interest. They proved mathematically that ARNNs have well-defined computational powers that extend the classical Universal Turing machine. Her initial publications on the computational power of Neural Networks culminated in a single-authored paper in Science and her monograph, "Neural Networks and Analog Computation: Beyond the Turing Limit".

In her Science paper, Siegelmann demonstrates how chaotic systems (that cannot be described by Turing computation) are now described by the Super-Turing model. This is significant since many biological systems not describable by standard means (e.g., heart, brain) can be described as a chaotic system and can now be modeled mathematically.

The theory of Super-Turing computation has attracted attention in physics, biology, and medicine. Siegelmann is also an originator of the Support Vector Clustering http://www.scholarpedia.org/article/Support_vector_clustering, a widely used algorithm in industry, for big data analytics, together with Vladimir Vapnik and colleagues. Siegelmann also introduced a new notion in the field of Dynamical Diseases, "the dynamical health" , which describes diseases in the terminology and analysis of dynamical system theory, meaning that in treating disorders, it is too limiting to seek only to repair primary causes of the disorder; any method of returning system dynamics to the balanced range, even under physiological challenges (e.g., by repairing the primary source, activating secondary pathways, or inserting specialized signaling), can ameliorate the system and be extremely beneficial to healing. Employing this new concept, she revealed the source of disturbance during shift work and travel leading to jet-lag and is currently studying human memory and cancer in this light.

Siegelmann has been active throughout her career in advancing and supporting minorities and women in the fields of Computer Science and Engineering. Through her career Siegelmann consulted with numerous companies, and has received a reputation for her practical problem solving capabilities. She is on the governing board of the International Neural Networks Society, and an editor in the Frontiers on Computational Neuroscience.

Papers

Cabessa, J.; Siegelmann, H. T. (2012). "The Computational Power of Interactive Recurrent Neural Networks". Neural Computation. 24 (4): 996–1019. doi:10.1162/neco_a_00263.

H.T. Siegelmann and L.E. Holtzman, "Neuronal integration of dynamic sources: Bayesian learning and Bayesian inference," Chaos: Focus issue: Intrinsic and Designed Computation: Information Processing in Dynamical Systems 20 (3): DOI: 10.1063/1.3491237, September 2010. (7 pages)

Nowicki, D.; Siegelmann, H.T. (2010). "Flexible Kernel Memory". PLOS One. 5: e10955. PMC 2883999 . PMID 20552013. doi:10.1371/journal.pone.0010955.

Olsen, M.M.; Siegelmann-Danieli, N.; Siegelmann, H.T. (2010). "Dynamic Computational Model Suggests that Cellular Citizenship is Fundamental for Selective Tumor Apoptosis". PLOS ONE. 5 (5): e10637. PMC 2869358 . PMID 20498709. doi:10.1371/journal.pone.0010637.

Pietrzykowski, A. Z.; Friesen, R. M.; Martin, G. E.; Puig, S.I.; Nowak, C. L.; Wynne, P. M.; Siegelmann, H. T.; Treistman, S. N. (2008). "Post-transcriptional regulation of BK channel splice variant stability by miR-9 underlies neuroadaptation to alcohol". Neuron. 59: 274–287. PMC 2714263 . PMID 18667155. doi:10.1016/j.neuron.2008.05.032.

Lu, S.; Becker, K.A.; Hagen, M.J.; Yan, H.; Roberts, A.L.; Mathews, L.A.; Schneider, S.S.; Siegelmann, H.T.; Tirrell, S.M.; MacBeth, K.J.; Blanchard, J.L.; Jerry, D.J. (2008). "Transcriptional responses to estrogen and progesterone in Mammary gland identify networks regulating p53 activity". Endocrinology. 149 (10): 4809–4820. PMC 2582927 . PMID 18556351. doi:10.1210/en.2008-0035.

Siegelmann, H.T. (2008). "Analog-Symbolic Memory that Tracks via Reconsolidation". Physica D: Nonlinear Phenomena. 237 (9): 1207–1214. doi:10.1016/j.physd.2008.03.038.

Roth, F.; Siegelmann, H.; Douglas, R. J. (2007). "The Self-Construction and -Repair of a Foraging Organism by Explicitly Specified Development from a Single Cell". Artificial Life. 13 (4): 347–368. doi:10.1162/artl.2007.13.4.347.

Leise, T.; Siegelmann, H.T. (2006). "Dynamics of a multistage circadian system". Journal of Biological Rhythms. 21 (4): 314–323. PMID 16864651. doi:10.1177/0748730406287281.

Loureiro, O.; Siegelmann, H. (2005). "Introducing an Active Cluster-Based Information Retrieval Paradigm". Journal of the American Society for Information Science and Technology. 56 (10): 1024–1030. doi:10.1002/asi.20193.

Ben-Hur, A.; Horn, D.; Siegelmann, H.T.; Vapnik, V. (2001). "Support vector clustering". Journal of Machine Learning Research. 2: 125–137.

Siegelmann, H.T.; Ben-Hur, A.; Fishman, S. (1999). "Computational Complexity for Continuous Time Dynamics". Physical Review Letters. 83 (7): 1463–1466. doi:10.1103/physrevlett.83.1463.

Siegelmann, H.T.; Fishman, S. (1998). "Computation by Dynamical Systems". Physica D. 120 (1–2): 214–235. doi:10.1016/s0167-2789(98)00057-8.

Siegelmann, H.T. (1995). "Computation Beyond the Turing Limit". Science. 238 (28): 632–637.

Partial List of Applications

Sivan, S.; Filo, O.; Siegelman, H. (2007). "Application of Expert Networks for Predicting Proteins Secondary Structure". Biomolecular Engineering. 24 (2): 237–243. doi:10.1016/j.bioeng.2006.12.001.

Eldar, S; Siegelmann, H. T.; Buzaglo, D.; Matter, I.; Cohen, A.; Sabo, E.; Abrahamson, J. (2002). "Conversion of Laparoscopic Cholecystectomy to open cholecystectomy in acute cholecystitis: Artificial neural networks improve the prediction of conversion". World Journal of Surgery. 26 (1): 79–85. doi:10.1007/s00268-001-0185-2.

Lange, D.; Siegelmann, H.T.; Pratt, H.; Inbar, G.F. (2000). "Overcoming Selective Ensemble Averaging: Unsupervised Identification of Event Related Brain Potentials". IEEE Transactions on Biomedical Engineering. 47 (6): 822–826. doi:10.1109/10.844236.

Karniely, H.; Siegelmann, H.T. (2000). "Sensor Registration Using Neural Networks". IEEE transactions on Aerospace and Electronic Systems. 36 (1): 85–98. doi:10.1109/7.826314.

Siegelmann, H.T.; Nissan, E.; Galperin, A. (1997). "A Novel Neural/Symbolic Hybrid Approach to Heuristically Optimized Fuel Allocation and Automated Revision of Heuristics in Nuclear Engineering". Advances in Engineering Software. 28 (9): 581–592. doi:10.1016/s0965-9978(97)00040-9.

Books

Neural Networks and Analog Computation : Beyond the Turing Limit, Birkhauser, Boston, December 1998 ISBN 0-8176-3949-7

She has also contributed 21 book chapters.

References

Hava Siegelmann Wikipedia

(Text) CC BY-SA