It makes sense to classify my interests into a few topics:
Extending the theory of probability: credal sets, full conditional measures, etc
Mostly I explore the
theory of sets of probability measures.
There is not a single, stable name for this theory: some people
use "theory of imprecise probabilities"; others say "theory of
credal sets", or "QuasiBayesian theory", or "theory of lower
expectations", or ... several other names. I'm a founding member of the
Society for
Imprecise Probability Theory and Applications; I also helped organize
some of the International Symposium for Imprecise Probabilities
and Their Applications (ISIPTA) and edited some of its proceedings.
I have developed graphbased models that represent sets of probability
measures over sets of variables; these are often called credal networks:
 F. G. Cozman.
Graphical Models for Imprecise Probabilities,
Journal of International Journal of Approximate Reasoning,
39(23):167184, 2005.
Preprint available.
 F. G. Cozman.
Credal networks, Artificial Intelligence Journal,
vol. 120, pp. 199233, 2000.
Preprint available.
Through the years I have produced many algorithms for inference with credal
networks, most of them with collaborators. A collection of still useful algorithms
can be found in

Jaime Shinsuke Ide, Fabio G. Cozman.
Approximate algorithms for credal networks with binary variables,
International Journal of Approximate Reasoning, v. 48, p. 275296, 2008.
Preprint available.

Cassio Polpo de Campos, Fabio Gagliardi Cozman.
Inference in credal networks through integer programming,
Fifth International Symposium on Imprecise Probability:
Theories and Applications, pp. 145154,
Prague, Czech Republic, 2007.
Preprint available.
 F. G. Cozman, C. P. de Campos, J. S. Ide, J. C. F. da Rocha.
Propositional and relational Bayesian networks associated
with imprecise and qualitative probabilistic assessments,
Conference on Uncertainty in Artificial Intelligence,
pp. 104111, AUAI Press, 2004.
Preprint available.
Additionaly, we have investigated combinations of graphbased modeling with
various concepts of independence, various kinds of assessments,
and with propositional/relational logic:

Cassio Polpo de Campos, Fabio G. Cozman, Jose Eduardo Ochoa Luna.
Assembling a consistent set of sentences in relational
probabilistic logic with stochastic independence,
Journal of Applied Logic, 7:137154, 2009.

Fabio G. Cozman, Cassio Polpo de Campos, Jose Carlos Ferreira da Rocha.
Probabilistic logic with independence,
International Journal of Approximate Reasoning, v. 49, p. 317, 2008.
Preprint available.

C. Polpo de Campos, F. G. Cozman.
Computing lower and upper expectations under
epistemic independence,
International Journal of Approximate Reasoning,
44(3):244260, 2007.
Preprint available.
 C. P. de Campos, F. G. Cozman.
Belief updating and learning in semiqualitative probabilistic
networks,
Conference on Uncertainty in Artificial Intelligence (UAI),
pp. 153160, Edinburgh, United Kingdom, 2005.
Preprint available.
Another interest of mine is the definition of independence between variables when
uncertainty is represented using sets of probability mesures. Some thoughts on this
can be found in:

Fabio G. Cozman.
Sets of probability distributions, independence, and convexity.
Synthese, 186(2):577600, 2012.
Preprint available.

F. G. Cozman, P. Walley.
Graphoid properties of epistemic irrelevance and
independence,
Annals of Mathematics and Artificial Intelligence,
45:173195, 2005.
Preprint available.
 F. G. Cozman.
Constructing Sets of Probability Measures Through Kuznetsov's
Independence Condition, Proceedings of the Second
International Symposium on Imprecise Probabilities and Their
Applications, pags. 104111, Ithaca, New York, United States, 2001.
Preprint available.
Here is an interesting piece on laws of large numbers for very general concepts of independence:

Fabio Gagliardi Cozman.
Concentration inequalities and laws of large numbers under epistemic
and regular irrelevance.
International Journal of Approximate Reasoning,
51:10691084, 2010.
Preprint available.
Still on concepts of independence, I have considered such concepts in the realm
of full conditional measures (that is, measures that extend standard probability
by adopting conditional probability as the primary object of interest, and hence
allowing conditioning on events of probability zero):

Fabio G. Cozman, Teddy Seidenfeld.
Independence for full conditional measures and their graphoid properties.
Foundations of the Formal Sciences VI,
Reasoning about Probabilities and Probabilistic Reasoning
pp. 129, College Publications, London, 2009.
Preprint available.
I have also looked at sequential decision making (that is, planning) under
uncertainty. An interesting idea is to merge "probabilistic" and "nondeterministic"
planning using the theory of sets of probabilities and Markov Decision Processes:

Felipe W. Trevizan, Fabio G. Cozman, Leliane N. de Barros.
Planning under Risk and Knightian Uncertainty,
International Joint Conference on Artificial Intelligence,
pp. 20232028, 2007.
Preprint available.
This work was followed up by papers that investigated computational issues:

Karina Valdivia Delgado, Leliane Nunes de Barros, Fabio Gagliardi
Cozman, Scott Sanner.
Using mathematical programming to solve Factored Markov Decision
Processes with Imprecise Probabilities.
International Journal of Approximate Reasoning, p. 200, 2011.
Preprint available.

Karina Valdivia Delgado, Scott Sanner, Leliane Nunes de Barros,
Fabio G. Cozman.
Efficient solutions to factored MDPs with imprecise transition probabilities,
19th International Conference on Automated Planning and Scheduling,
pp. 98105, Thessaloniki, Greece, 2009.
Preprint available.
Still on sequential decision making, here are publications that deal with
decision trees and sets of probability measures:

Daniel Kikuti, Fabio G. Cozman, Ricardo Shirota Filho.
Sequential decision making with partially ordered preferences.
Artificial Intelligence, 175(78):13461365, 2011.
Preprint available.
 D. Kikuti, F. G. Cozman, C. P. de Campos.
Partially ordered preferences in decision trees: computing
strategies with imprecision in probabilities,
IJCAI Workshop on Advances in Preference Handling,
Edinburgh, United Kingdom, 2005.
Preprint available.
PLEASE also note that an errata has been produced,
correcting a few mistakes in the paper!
And here is an older effort, looking at the problem of sequentialdecision
making associated with observations:
 F. Cozman; E. Krotkov.
QuasiBayesian Strategies for Efficient Plan
Generation: Application to the Planning to Observe Problem,
Proc. Twelfth Conference Uncertainty in Artificial Intelligence,
pp. 186193, 1996.
Preprint available.
Changing the subject a little,
here is a paper that summarizes old work on calculations with sets
of probability measures:
 F. G. Cozman.
Computing posterior upper expectations,
International Journal of Approximate Reasoning,
vol. 24, pp. 191205, 2000.
Preprint available.
Finally, a long time ago I explored with Lonnie Chrisman the possibility
of learning convex sets of probability from data; that old work
has been picked up by Terry Fine and his collaborators.
Probabilistic reasoning: knowledge representation, and machine learning
I am quite interested in probabilistic models in knowledge representation
and machine learning.
Relational Bayesian networks and description logics
I have been interested, together with several students,
in languages for knowledge representation and machine
learning that combine graphbased modeling, in particular Bayesian networks,
with logical structures, in particular description logics. Perhaps the most
interesting language we have developed is

Fabio G. Cozman, Rodrigo Polastro.
Complexity analysis and variational inference for
interpretationbased probabilistic description logics,
Conference on Uncertainty in Artificial Intelligence, 2009.
Preprint available.
We have also looked at algorithms that learn descriptions in this
language from data:

Jose Eduardo Ochoa Luna, Kate C. Revoredo, Fabio Gagliardi Cozman.
Learning probabilistic description logics: A framework and algorithms,
Advances in Artificial Intelligence  10th Mexican International Conference on
Artificial Intelligence (MICAI2011), Lecture Notes in Artificial Intelligence 7094
Part I, pp. 2839, Springer, 2011.
Preprint available.
We have also looked at a variety of applications, from social network
analysis, to localization in mobile robotics, to spatial reasoning. Here
are some representative publications:

Jose Eduardo Ochoa Luna, Kate C. Revoredo, Fabio Gagliardi Cozman.
An Experimental Evaluation of a Scalable Probabilistic Description Logic Approach for Semantic
Link Prediction.
8th International Workshop on Uncertainty Reasoning for the Semantic Web (URSW 2012),
pp. 6374, 2012.
Preprint available.

Rodrigo B. Polastro, Fabiano E. Correa, Fabio G. Cozman, J. Okamoto Jr.
Semantic mapping with a probabilistic description logic.
Lecture Notes in Artificial Intelligence, volume 6404,
Advances in Artificial Intelligence  SBIA 2010, pp. 6271, 2010.
Preprint available.

Paulo E. Santos, Fabio G. Cozman, Valquiria F. Pereira, B. Hummel.
Probabilistic logic encoding of spatial domains.
International Workshop on Uncertainty in Description Logics, 2010.
Preprint available.
Semisupervised learning
One of the most important situations where we make decisions
and use our beliefs and sensory information is when we
classify data.
I have worked with classification problems where one
must build ("learn") a classifier using observed data that can
be labeled or unlabeled;
that is, the data points themselves may be classified or not.
This is often called semisupervised learning.
Relevant publications are:
 I. Cohen, F. G. Cozman, N. Sebe, M. C. Cirelo, T. S. Huang.
Semisupervised learning of classifiers: Theory, algorithms,
and their application to humancomputer interaction,
IEEE Transactions on Pattern Analysis and Machine Intelligence,
26(12):15531568, 2004.
Preprint
available.

F. G. Cozman, I. Cohen,
Risks of semisupervised learning,
in Olivier Chapelle, Bernhard Scholkopf, Alexander Zien
(editors),
SemiSupervised Learning, pp. 5570, 2006.
Preprint available.
 F. G. Cozman, I. Cohen, M. C. Cirelo. Semisupervised
learning of mixture models, International Conference
on Machine Learning, pp. 99106, 2003.
Preprint available.
 I. Cohen, N. Sebe, F. G. Cozman, M. C. Cirelo, T. S. Huang.
Learning Bayesian network classifiers for facial expression
recognition using both labeled and unlabeled data, IEEE
Conference on Computer Vision and Pattern Recognition, 2003.
Preprint available.
This was joint work with
Ira Cohen,
at HP Labs Palo Alto. Other people at HP Labs
contributed a lot, particularly Marsha Duro and Alex Bronstein.
JavaBayes and Bayesian networks
I developed, between 1996 and 2002, the
JavaBayes
system, a general purpose inference engine for graphical models.
The engine computes posterior
probabilities and expectations for probabilistic models represented as
directed acyclic graphs. The system is distributed freely
(under the GNU license) in the spirit of fostering teaching and
research. JavaBayes
is now used in many university and research labs around the
world. A summary can be found in:
 F. G. Cozman.
The JavaBayes system,
The ISBA Bulletin, vol. 7, n. 4, pp. 1621, 2001
(invited publication without referreing process).
In the process of putting together JavaBayes, I have
developed a very general, yet easy to understand, inference
algorithm for Bayesian networks. The method is suited for
teaching due to its simplicity. You can get it:
While
JavaBayes
is a complete system, with graphical interface, parsers, etc,
the EBayes
package was an early effort to produce a lightweight Bayesian network
engine that is appropriate to the growing market of embedded
devices. A much more complete algorithm for probabilistic inference
under time and space constraints has been developed with my former
student Fabio Ramos:
 F. T. Ramos, F. G. Cozman.
Anytime anyspace probabilistic inference,
International Journal of Approximate Reasoning,
38:5380, 2005.
Preprint available.
Still on Bayesian networks, I have worked with my former student
Jaime S. Ide on methods for generating random Bayesian networks:
 J. S. Ide, F. G. Cozman, F. T. Ramos.
Generating random Bayesian networks with constraints
on induced width,
European Conference on Artificial Intelligence (ECAI),
pp. 323327, IOS Press, Amsterdan, 2004.
Preprint available.
Robotics and Computer Vision: Teleoperation, mobile robots, automated orthosis...
Third, I was involved for a long time, in one way or another,
with robotic devices, mostly with mobile robots. I have not worked on these topics
for a while.
My involvement with robotics started quite a while ago.
Right after my undergraduate course,
I took a Master of Engineering
in Brazil, and worked in the first Brazilian mobile robot, called Ariel.
We produced a complete system, from the mechanical structure to the
planning software; the result was very impressive and we ended up
showing it off in the Jornal da Globo (Brazil's second most important
TV news source at the time). Unfortunately, that material is not online.
Here are two representative papers, perhaps of historic value:
 F. G. Cozman; P. E. Miyagi. Trajectory Controller for a Mobile
Robot using Optimal Control, XI Congresso Brasileiro de Engenharia
Mecânica, 3:537540, São Paulo, SP Brasil, 1991.
 J. C. Adamowski; M. G. Simões; F. G. Cozman. Desenvolvimento
de um Robô Móvel, VIII
Congresso Brasileiro de Automática,
Belém, 1990; selected for IV Congreso Latinoamericano de
Control Automatico, Puebla Mexico, 1990; also presented at
IV Congresso Nacional de
Automação
Industrial, pp. 209212, São Paulo, SP Brasil, 1990.
I worked, for two years, in the Lunar Rover project
during my PhD years at Carnegie Mellon.
My main contribution to the Lunar Rover project was the
Viper
system, a piece of technology
that was used in the Atacama desert for tests.
The Viper system,
estimates position from a stream of images, by matching images
to a previously constructed map of the environment.
The estimator builds an occupancy map for the position
of the robot; the catch is that the occupancy maps actually represents
a full density ratio familiy of distributions which generate both
the estimates and the confidence on the estimates.
The system is described in:
 F. G. Cozman, E. Krotkov, C. E. Guestrin.
Outdoor Visual Position Estimation for Planetary Rovers,
Autonomous Robots, vol. 9, pp. 135150, 2000.
There is also a description of an old version of the Viper system in:
 F. Cozman; E. Krotkov.
Automatic Mountain Detection
and Pose Estimation for Teleoperation of Lunar Rovers,
Proc. of the International Conference on Robotics
and Automation, pp. 24522457, Albuquerque, New Mexico, 1997.
Also published in
Experimental Robotics V,
Lecture Notes in Control and Information Sciences 232,
pp. 207215,
Alicia Casals e Anibal T. de Almeida (eds.),
Barcelona, Spain, June (1518) 1997.
The vision algorithms developed for the Viper system
are reported in the following papers.
 C. Guestrin; F. G. Cozman; E. Krotkov.
Fast Software Image Stabilization with Color Registration,
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS),
pp. 1924, Victoria, Canada, October, 1998.
 C. Guestrin; F. G. Cozman; M. G. Simões.
Industrial Applications of Image Mosaicing and Stabilization,
Second International Conference on Knowledgebased Intelligent
Electronic Systems, pp. 174183, Adelaide, Australia, April 1998.
During a few years at CMU I worked with the
Ratler
robot. We actually had it rolling for some fifty
kilometers in our outdoor tests; you can take a look
at the following paper.
 R. Simmons; E. Krotkov; L. Chrisman; F. Cozman; R. Goodwin;
M. Hebert; L. Katragadda; S. Koenig; G. Krishnaswamy; Y. Shinoda; W.
Whittaker; and P. Klarer.
Experience with Rover Navigation for LunarLike Terrains,
Proceedings of the Conference on Intelligent Robots
and Systems (IROS), pages 441446, 1995.
I also worked on a variety of other problems. For instance,
once I wrote a line linker based on the Akaike Information Criterion,
that was distributed in the web. Another example was the investigation of
celestial data as a source of position estimates for mobile robots:
 F. Cozman; E. Krotkov. Robot
Localization using a Computer Vision Sextant,
International Conference on Robotics and Automation, pages 106111,
Nagoya, Japan, May 1995.
I was interested for some time in the problem of calculating
bounds for dynamical systems; there is a huge literature
in this area. I have published some work on the specific
topic of manipulating ellipsoidal models of error in Robotics:
 F. Cozman; E. Krotkov. Truncated
Gaussians as Tolerance Sets,
Fifth Workshop on Artificial Intelligence and Statistics, Fort Lauderdale
Florida, 1995.
Yet another example was the study of atmospheric scattering as a
clue for depth in outdoor environments; as far as I know, the first
study of scattering in the context of image understanding.
 F. Cozman; E. Krotkov. Depth from Scattering,
Proceedings of the IEEE Conference on Computer Vision
and Pattern Recognition, Puerto Rico, June, 1997.
Finally, during 20002002, I participated in an effort to develop
devices that can help the disabled walk with less effort and
discomfort. The project started from interactions with doctors
and engineers at the Associação
de Assistência a Criança Defeituosa and was
supported by FAPESP.
A former advisee involved with this project, Marco Ackermann,
received the prize of Best Master Thesis in Mechanical Engineering
in Brazil 2003, granted by the Brazilian Association for the
Mechanical Sciences (ABCM), for this work. You can read about it:

Marko Ackermann, Fabio Gagliardi Cozman.
Automatic knee flexion in lower limb orthoses.
Journal of the Brazilian Society of Mechanical Sciences
and Engineering, 31(4):305311, 2009.
Preprint available.
Since then I have occasionally returned to robotics and computer vision
when dealing with uncertain reasoning and knowledge representation; my focus has
been on the latter topics and not so much on the robots themselves.
For instance:

Rafael A. M. Goncalves, Diego R. Cueva, Marcos R. PereiraBarretto, Fabio G. Cozman.
A model for inference of emotional state based on facial expressions,
Journal of the Brazilian Computer Society, 2012 (Online first).
Preprint available.

Valquiria Fenelon Pereira, Paulo E. Santos, Hannah M. Dee and Fabio G. Cozman.
Reasoning about shadows in a mobile robot environment.
Applied Intelligence, 2012 (Online first).
Preprint available.

Tiago Mato, Yannick P. Bergamo, Valdinei F. da Silva, Fabio G. Cozman, Anna H. Reali Costa.
Simultaneous Abstract and Concrete Reinforcement Learning,
Symposium on Abstraction, Reformulation, and Approximation, pp. 8289,
Parador de Cardona, July 2011.
Preprint available.
fgcozman@usp.br