Gadi Pinkas
Dr. Pinkas can still be reached at pinkas@cs.wustl.edu.
Education
1989-1992 Washington University, St. Louis (D.Sc.)
1983-1985 Hebrew University, Jerusalem, ISRAEL (M.S.)
1977-1982 Bar Ilan University, Ramat Gan, ISRAEL (B.S.)
Relevant Experience
1992-present Project Leader, Amdocs Inc., St. Louis and Tel Aviv, ISRAEL
1992-1995 Research Affiliate, Center for Semantic Control and Optimization,Washington University, St. Louis
1986-1989 Project Leader, Amdocs Inc., St. Louis
1981-1986 Project Manager, Cohatal, Ltd., Jerusalem, ISRAEL
1975-1981 Programmer and Designer, IDF, ISRAEL
Awards
The 1985 Information Processing Association of Israel prize for the work
"A linguistic information retrieval system."
Thesis
Logical Inference in Symmetric Connectionist Networks
Committee: R. Loui (Advisor), D. Touretzky (Carnegie Mellon, external advisor),
T. D. Kimura, S. C. Kwasny, W. E. Ball, E. Y. Rodin (Systems Science and Mathematics)
Won competition for departmental nomination for ACM dissertation of the
year award (declined while seeking independent publication).
Professor Rodin at the defense: "This thesis reminds me of the thesis of Weierstrass."
Publications
Pinkas, G.
"CONTEXT -- a natural language full-text retrieval system,"
Proc. of the Conference on User Oriented content-Based
Text and Image Handling (RIAO-88), 1988.
Pinkas, G.
"A linguistic information retrieval system,"
Maase Hoshev, the IPA Journal 12 : 6, 1985. Also
Proc. of the Israel National Conf. on Data Processing,
1985.
Pinkas, G.
"Energy minimization and the satisfiability of propositional logic,"
in Proc. of the Connectionist Models School, D. Touretzky, J. Elman,
T. Sejnowskiand G. Hinton (eds.), Morgan Kaufmann, 1990. Also Presented
at the Second Midwest AI and Cognitive Science Conference, 1990.
Pinkas, G.
"The equivalence of connectionist energy minimization and propositional calculus satisfiability,"
technical report, WU CS 90-03,
1990.
Pinkas, G.
"Symmetric neural networks and propositional logic satisfiability,"
Neural Computation 3 : 2,
1991.
Pinkas, G.
"First-order logic proofs using connectionist constraint relaxation,"
technical report, WU CS 91-54,
1991.
Pinkas, G.
"Symmetric neural nets and propositional logic satisfiability,"
technical report, WU CS 91-02,
1991.
Pinkas, G.
"Propositional non-monotonic reasoning and inconsistency in symmetric neural networks,"
Proc. IJCAI, 1991.
Also technical report, WU CS 91-03,
1991.
Pinkas, G and R. Loui.
"Reasoning from inconsistency : a taxonomy and a connectionist approach,"
Proc. Knowledge Representation and Reasoning (KR92),
1992.
Also technical report, WU CS 91-27,
1991.
Pinkas, G.
"Representation and learning of propositional knowledge in symmetric connectionist networks,"
technical report, WU CS 91-52,
1991.
Pinkas, G.
"Converting binary threshold networks into equivalent symmetric networks,"
technical report, WU CS 91-31,
1991.
Pinkas, G. and Rina Dechter.
"An improved connectionist activation function for energy minimization,"
Proceedings of AAAI-92.
Also Information and Computer Science, University of California, Irvine, technical report 92-62,
1992.
Pinkas, G.
"Logical inference in symmetric connectionist networks,"
Doctoral Dissertation, Washington University, Dept. of Computer Science,
1992.
Pinkas, G.
"Propositional logic, nonmonotonic reasoning and symmetric networks:
bridging the gap between symbolic and connectionist knowledge
representation,"
in Neural Networks for Knowledge Representation and Inference,
D. Levin and M. Aparicio (eds.), Erlbaum, 1992.
Pinkas, Gadi
"Constructing syntactic proofs in symmetric networks,"
Advances in Neural Information Processing Systems 4 (NIPS 91),
1992.
Pinkas, G.
"A fault tolerant connectionist architecture for construction of logic proofs,"
in AI and Neural Networks, V. Honavar and L. Uhr, eds., Academic Press, 1994.
Also technical report, WU CS 93-10,
1993.
Pinkas, G.
"Representing and learning propositional logic in symmetric connectionist networks,"
technical report, WU CS 93-09,
1993.
Pinkas, G.
"Logical inference in symmetric connectionist networks,"
technical report, WU CS 93-11,
1993.
Pinkas, G.
"Representing propositional logic and searching for satisfiability in connectionist networks,"
Machine Intelligence and Pattern Recognition 14,
1994.
Pinkas, G. and R. Dechter.
"Improving connectionist energy minimization,"
Journal of AI Research (JAIR) 3,
1995.
Pinkas, G.
"Reasoning, nonmonotonicity and learning in connectionist networks that capture propositional knowledge,"
Artificial Intelligence 77 : 2,
1995.