The Official MathGeek Blog

Knock Knock Jokes

Here's an interesting development: A couple of University of Cincinnati researchers have developed an artificial-intelligence program that can understand knock-knock jokes.

Knock-knock jokes, of course, frequently rely on a pun, as with: "Knock, Knock. Who is there? Wendy. Wendy who? Wendy last time you took a bath?" Humans intuitively understand when a pun is being made, because we're able to notice that a crucial word in the punchline is being misused semantically -- it's an incorrect meaning -- and is riding along purely because it sounds like the correct word. But computers have huge, huge problems with grasping semantics and homonyms.

So to create their knock-knock joke 'bot, the researchers programmed their AI with a big list of homonyms and their various meanings -- including, significantly, a lot of proper names (like "Wendy") because a lot of knock-knock jokes rely on proper names. Then whenever the AI reads a new knock-knock joke, it identifies the crucial "joke" word, then pings its database of homonyms to see if any of the words' rival meanings "fits" the joke. If it does, the bot flags the joke as "funny."  More...

Publications by Eric Hall

1. Wise, G. L., and E. B. Hall, Counterexamples in Probability and Real Analysis, Oxford University Press, New York, 1993, (ISBN 0–19–507068–2). (Review)
2. Hall, E. B., Alan Wessel, and Gary Wise, "Some Aspects of Fusion in Estimation Theory," Selected Papers on Sensor and Data Fusion, SPIE Milestone Series, Volume MS 124, edited by Firooz A. Sadjadi, 1996, pp. 62–64 (ISBN 0–8194–2265–7).
3. Hall, E. B. and G. L. Wise, "An Algebraic Aspect of Linear System Theory," IEEE Transactions on Circuits and Systems, Vol. 37, No. 5, May 1990, pp. 651–653.
4. Hall, E. B. and G. L. Wise, "A Curiosum Concerning Discrete Time Convolution," IEEE Trans. on Acoustics, Speech, and Signal Processing, Vol. 38, No. 6, June 1990, pp. 1058–1059.
5. Wessel, A. E., E. B. Hall and G. L. Wise, "Importance Sampling Via a Simulacrum," The Journal of the Franklin Institute, Vol. 327, No. 5, 1990, pp. 771–783.
6. Wise, G. L. and E. B. Hall, "A Note on the Distribution of the Determinant of a Random Matrix," Statistics and Probability Letters, Vol. 11, No. 2, pp. 147–148, February 1991.
7. Hall, E. B., A. E. Wessel and G. L. Wise, "Some Aspects of Fusion in Estimation Theory," IEEE Transactions on Information Theory, Vol. 37, No. 2, pp. 420–422, March 1991.
8. Hall, E. B. and G. L. Wise, "On Optimal Estimation with Respect to a Large Family of Cost Functions," IEEE Transactions on Information Theory, Vol. 37, No. 5, pp. 691–693, May 1991.
9. Wise, G. L. and E. B. Hall, "A Comment on the Finite Memory of Stochastic Processes," IEEE Transactions on Signal Processing, Vol. 40, No. 9, p. 2368, September 1992.
10. Hall, E. B. and G. L. Wise, "A Result on Multidimensional Quantization," Proceedings of the American Mathematical Society, June 1993, Vol. 118, No. 2, pp. 609–613.
11. Wise, G. L. and E. B. Hall, "On Mutual Independence," Statistics and Probability Letters, Vol. 17, No. 5, 1993, pp. 395–398.
12. Hall, E. B., "A Note on the Convergence of Quantizers," Mathematical Problems in Engineering, Vol. 2, pp. 57–61, 1996.
13. Hall, E. B. and G. L. Wise, "Simultaneous Optimal Estimation Over a Family of Fidelity Criteria," Proceedings of the Twenty–First Annual Conference on Information Sciences and Systems, Baltimore, Maryland, March 25–27, 1987, pp. 65–70.
14. Wessel, A. E., E. B. Hall and G. L. Wise, "Some Comments on Importance Sampling," Proceedings of the Twenty–Second Annual Conference on Information Sciences and Systems, Princeton, New Jersey, March 16–18, 1988, pp. 325–329 (invited).
15. Hall, E. B., A. E. Wessel and G. L. Wise, "On Fusion in Estimation Theory," Proceedings of the Twenty–Sixth Annual Allerton Conference on Communication, Control, and Computing, Monticello, Illinois, September 28–30, 1988, pp. 599–608.
16. Hall, E. B. and G. L. Wise, "Conditioning: A Critical Review," Proceedings of the 23rd Annual Conference on Information Sciences and Systems, Baltimore, Maryland, March 22–24, 1989, pp. 271–276.
17. Hall, E. B. and G. L. Wise, "Some Comments on Filtering in Estimation Theory," Proceedings of the 32nd Midwest Symposium on Circuits & Systems, Urbana, Illinois, August 14–16, 1989, pp. 1162–1165.
18. Hall, E. B. and G. L. Wise, "On an Aspect of Optimal Nonlinear Estimation," Proceedings of the Twenty–Seventh Annual Allerton Conference on Communication, Control, and Computing, Monticello, Illinois, September 27–29, 1989, pp. 692–701.
19. Hall, E. B. and G. L. Wise, "Some Pathologies Associated with Maximum Likelihood Estimation," Proceedings of the Twenty–Fourth Annual Conference on Information Sciences and Systems, Princeton, New Jersey, pp. 11–16, March 21–23, 1990.
20. Hall, E. B. and G. L. Wise, "An Analysis of Convergence Problems in Kalman Filtering which Arise from Numerical Effects," Proceedings of the 33rd Midwest Symposium on Circuits and Systems, Calgary, Alberta, Canada, pp. 315–318, August 12–14, 1990.
21. Hall, E. B. and G. L. Wise, "Some Comments on Stochastic Calculus," Proceedings of the 1991 Conference on Information Sciences and Systems, Baltimore, Maryland, March 20–22, 1991.
22. Hall, E. B. and G. L. Wise, "Some Aspects of Multidimensional Convolution," Proceedings of the 1991 International Conference on Acoustics, Speech, and Signal Processing, Toronto, Ontario, Canada, May 14–17, 1991, pp. 2317–2320.
23. Hall, E. B. and G. L. Wise, "On a Failure of Multidimensional Quantization," Proceedings of the 1991 International Conference on Acoustics, Speech, and Signal Processing, Toronto, Ontario, Canada, May 14–17, 1991, pp. 1921–1924.
24. Hall, E. B. and G. L. Wise, "Decentralized Estimation with Nontraditional Fidelity Criteria and Corrupted Estimates," Proceedings of the 1992 Conference on Information Sciences and Systems, Princeton, New Jersey, March 18–20, 1992.
25. Wise, G. L. and E. B. Hall, "Estimation of a Random Variable Based on Multidimensional Data," Proceedings of the 1992 International Conference on Acoustics, Speech, and Signal Processing, Vol. 3, pp. 317–320, San Francisco, California, March 23–26, 1992.
26. Joudeh, I. and E. B. Hall, "Delay Analysis for Packet Trains in Computer Communications using Single Server Queues," Proceedings of the 1992 National Telesystems Conference, Washington, D.C., May 19–20, 1992, pp. 13-7–13-14.
27. Bienn, M. and E. B. Hall, "Symbol Synchronization over a Frequency Nonselective Fading Channel," Proceedings of the IEEE Wichita Conference on Communications, Networking and Signal Processing, Wichita, Kansas, April 26–27, 1994.
28. Hall, E. B., "On the Modeling of Quantization Effects via Points of Accuracy," Proceedings of the 33rd IEEE Conference on Decision and Control, Lake Buena Vista, Florida, pp. 2522–2523, Dec. 14–16, 1994.
29. Bienn, M. and E. B. Hall, "A Decision Theoretic Approach to Symbol Synchronization over a Frequency Nonselective Fading Channel," Proceedings of the 34th Conference on Decision and Control, New Orleans, Louisiana, pp. 2070–2075, Dec. 13–15, 1995.
30. Hall, E. B., "The Composite Classification Problem in Optical Information Processing," NASA/ASEE Summer Faculty Fellowship Program – 1994, Vol. 1, Richard Bannerot and Donn G. Sickorez, eds., NASA Contractor Report 188410, Grant NGT 44–001–800, July 1995, Chapter 13.
31. Aliyazicioglu, Z. and E. B. Hall, "Optimal Decision for Distributed Detection Systems," Proceedings of the Second World Conference on Integrated Design and Process Technology, Austin, Texas, pp. 121-127, December 1-4, 1996.

Maximum Likelihood Estimation

The theoretical statistical community has long been aware of the many problems associated with the concept of maximum ljkelihood estimation. In fact, Sir Ronald Fisher's 1922 revival of the maximum likelihood method is seen by many to have been a historical retrogression since the method had been considered and rejected years earlier by none other than Carl Friedrich Gauss. The popularity of maximum likelihood estimation today is due primarily, in the view of Le Cam, to Fisher's "propaganda." In fact, Le Cam states that "In view of Fisher's vast influence, it is perhaps not surprising that the presumed superiority of the method is still for many an article of faith promoted with religious fervor. This state of affairs remains, in spite of a long accumulation of evidence to the effect that maximum likelihood estimators are often useless, or grossly misleading." It is precisely this evidence with which this paper is concerned, since, judging by its popularity, many of the problems associated with the maximum likelihood method appear to have been overlooked by many in the area of information sciences and systems. We begin with a definition of the maximum likelihood method. More...

A note on the convergence of quantizers

Several questions concerning the convergence of quantized estimators are considered based upon points of accuracy. Examples are considered in which quantizers converge and in which they fail to converge, and points of accuracy are shown to provide explanations for both types of behavior.  Read the Article Here.