IEEE Signal Processing Magazine - January 2018 - 171

information measures as optimal inference problems can be used to derive
learning algorithms, such as in [6], as
well as estimates of information measures [5], [11].

Author
Osvaldo Simeone (osvaldo.simeone@
kcl.ac.uk) received his M.Sc. degree
(with honors) and his Ph.D. degree,
both in information engineering, from
Politecnico di Milano, Italy, in 2001
and 2005, respectively. He is a professor of information engineering with
the Centre for Telecommunications
Research in the Department of In formatics of King's College London.
From 2006 to 2017, he was a faculty
member with the Electrical and Computer
Engineering Department at the New
Jersey Institute of Technology. His
research interests include wireless com-

FROM THE EDITOR

munications, information theory, optimization, and machine learning. He is
a corecipient of the 2017 JCN Best
Paper Award, the 2015 IEEE Communication Society Best Tutorial
Pa per Award, and the Best Paper
Awards of IEEE SPAWC 2007 and
IEEE WRECOM 2007. He was awarded a European Research Council Consolidator Grant in 2016. He is a Fellow
of the IEEE.

[5] X. Nguyen, M. J. Wainwright, and M. I. Jordan,
"Estimating divergence functionals and the likelihood
ratio by convex risk minimization," IEEE Trans.
Inform. Theory, vol. 56, no. 11, pp. 5847-5861, Nov.
2010.

References

[9] J. C. Duchi, Lecture Notes for Statistics 311/
Electrical Engineering 377. Stanford, CA.

[1] T. M. Cover and J. A. Thomas, Elements of
Information Theory. Hoboken, NJ: Wiley, 2012.
[2] C. M. Bishop, Pattern Recognition and Machine
Learning. New York: Springer, 2006.
[3] I. Csiszár, "Axiomatic characterizations of information measures," Entropy, vol. 10, no. 3, pp. 261-
273, Sept. 2008.
[4] P. Grünwald and P. Dawid, "Game theory, maximum entropy, minimum discrepancy and robust
Bayesian decision theory," Ann. Statist., vol. 32, no.
4, pp. 1367-1433, 2004.

[6] S. Nowozin, B. Cseke, and R. Tomioka, "f-GAN:
Training generative neural samplers using variational divergence minimization," in Proc. Neural
Information Processing Systems Conf., Barcelona,
Spain, 2016, pp. 4240-4248.
[7] C. E. Shannon, "A mathematical theory of communication," Bell Syst. Tech. J., vol. 27, no. 3, pp.
379-423, July 1948.
[8] J. Jiao, T. A. Courtade, A. No, K. Venkat, and T.
Weissman, "Information measures: The curious case
of the binary alphabet," IEEE Trans. Inform. Theory,
vol. 60, no. 12, pp. 7616-7626, Dec. 2014.

[10] M. H. DeGroot, "Changes in utility as information," Theory Decis., vol. 17, no. 3, pp. 287-303, Nov.
1994.
[11] V. Berisha, A. Wisler, A. Hero, and A. Spanias,
"Empirically estimable classification bounds based on
a nonparametric divergence measure," IEEE Trans.
Signal Proc., vol. 64, no. 3, pp. 580-591, Feb. 2016.
[12] M. Arjovsky, S. Chintala, and L. Bottou,
"Wasserstein GAN," arXiv Preprint, arXiv:1701.07875,
Jan. 2017.

(continued from page 4)

objective of creating content that can
be used to promote the wealth of signal
processing applications in lower-level
undergraduate courses. I believe such
articles would provide additional motivation for students and professors in signal
processing. Content from industry will
enhance the offerings of SPM, making
it more valuable to the readership and
enhancing the underlying industrial support of contributors.
Standards and commercialization are
important in many areas of signal processing, which are not covered extensively in SPM. For example, communication
standards are driving the development
of fifth-generation signal processing
algorithms while the commercialization of virtual reality by several firms is
encouraging further development in low-

power multimedia signal processing for
consumers. I envision more columns or
articles where the authors provide a toplevel view of the standards development
including what is being standardized,
the time line, how the meetings are conducted, and where information is shared.
This will be of great value for anyone not
attending the meetings. In terms of commercialization, I foresee contributions
related to emerging consumer products
(explaining the signal processing connection, not just blanket marketing) or highlights of different technology start-ups.
Content in these new areas will enhance
the appeal of the magazine.
Signal processing can be funny. Many
graduate students remember Ph.D. comics or the cartoons in signal processing
books about the terrors on convolution.
IEEE SIGNAL PROCESSING MAGAZINE

January 2018

I will work to develop some lighthearted
content in SPM, composed of contributions from the readership. These could
take the form of short articles or cartoons, including plays on words, the trials
of graduate school, or even fun puzzles.
Such contributions will provide a hook
that will keep readers looking through
every issue and will help create a fun culture that brings people together.
SPM is a magazine for all of us. I look
forward to your feedback and ideas.

SP
171

Table of Contents for the Digital Edition of IEEE Signal Processing Magazine - January 2018