jsims@mwunix.mitre.org (James Sims) (01/23/91)
Can someone point me to a (hopefully) existing comparision between various learning techniques - GA, Neural Nets, Adaptive Control Theory, learning in rulebases, etc in the literature? If not, are there survey references that describe the strengths/weaknesses of each of these approaches individually? reply by email to sims@starbase.mitre.org, I'll summarize to the net. thanks in advance, jim
tgd@tesla.orst.EDU (Tom Dietterich) (01/25/91)
Many people have been comparing alternative learning algorithms. In IJCAI-89 (International Joint Conference on Artificial Intelligence), there were three papers published: Weiss, S., and Kapouless, I., (1989). An empirical comparison of pattern recognition, neural nets, and machine learning classification methods. {\it IJCAI-89: Eleventh International Joint Conference on Artificial Intelligence}. 781--787. Mooney, R., Shavlik, J., Towell, G., and Gove, A. (1989). An experimental comparison of symbolic and connectionist learning algorithms. {\it IJCAI-89: Eleventh International Joint Conference on Artificial Intelligence}. 775--80. Fisher, D. H., and McKusick, K. B., (1988). An empirical comparison of ID3 and back-propagation. {\it IJCAI-89: Eleventh International Joint Conference on Artificial Intelligence}. 788--793. At the last three meetings of NIPS (Neural Information Processing Systems), there have been talks comparing a wide range of methods. Gish, S. L. and Blanz, W. E. 1990. Comparing the performance of connectionist and statistical classifiers on an image segmentation problem. In D. Touretzky (Ed.), Advances in Neural Information Processing Systems (Vol. 2). San Mateo, CA: Morgan Kauffman. 614--621. Atlas, L., Cole, R., Connor, J., El-Sharkawi, M., Marks, R. J. II, Muthusamy, Y., Barnard, E. 1990. Performance comparisons between backpropagation networks and classification trees on three real-world applications. In D. Touretzky (Ed.), Advances in Neural Information Processing Systems (Vol. 2). San Mateo, CA: Morgan Kauffman. 622-629. Tsoi, A. C., and Pearson, R. A. 1991. Comparison of three classification techniques: CART, C4.5, and multi-layer perceptrons. To appear in D. Touretzky (Ed.), Advances in neural Information Processing Systems (Vol 3.), San Mateo, CA: Morgan Kaufmann. Ng, K. and Lippmann, R. P. 1991. A comparative study of the practical characteristics of neural network and conventional pattern classifiers. To appear in D. Touretzky (Ed.), Advances in neural Information Processing Systems (Vol 3.), San Mateo, CA: Morgan Kaufmann. Finally, my students and I have contributed to this literature as well: Dietterich, T. G., Hild, H., Bakiri, G. (1990) A comparative study of ID3 and backpropagation for English text-to-speech mapping. {\it Proceedings of the Seventh International Conference on Machine Learning} (pp. 24--31). Austin, TX: Morgan Kaufmann. I'm sure I've missed some papers. These are just the ones that I could find in a few minutes. Thomas G. Dietterich Department of Computer Science Dearborn Hall, 303 Oregon State University Corvallis, OR 97331-3102 503-737-5559