UCBI 2015: workshop on Unconventional Computing for Bayesian Inference/IROS 2015

This workshop will address unconventional computing for Bayesian inference, with keynote speakers on Bayesian inference for autonomous robots, and insights form computational biology, as well as presentations of submitted works, aiming to encourage these Bayesian and unconventional computing approaches to the IROS community.

Invited Speakers

Jacques Droulez, "Bayesian computing in biology"

Pierre Bessičre, "Bayesian Programming for Robotics"

Joćo Filipe Ferreira, "Probabilistic Approaches for Robotic Perception"

Jorge Dias, "Probabilistic Approaches for Robotic Perception"


The workshop on unconventional computing for Bayesian inference invites submissions of full papers or extended abstracts for poster presentation. Topics include, but are not strictly limited to:

* Low power computing solutions for Bayesian inference

* Parallel architectures and unconventional computing for Bayesian inference

* Stochastic computing for Bayesian inference

* Autonomous robots performing Bayesian inference with limited resources

* Insights from computational biology for Bayesian computing

* Bayesian programming for robotics

* Bayesian models for robotic perception and cognition

Submission should follow the conference format , 6 page for the full papers and 2 page for extended abstracts, and submitted via easychair For any help or further details please contact

Important dates

Submission deadline: July 6

Notification of acceptance: July 31

Camera ready submission: September 1

Workshop day: September 28


Contemporary robots and other cognitive artifacts are not available to autonomously operate in complex environments. The major reason for this failure is the lack of cognitive systems able to efficiently deal with uncertainty when behaving in real world situations.

One of the challenges of robotics is endowing devices with adequate computational power to dwell in uncertainty and decide with incomplete data, with limited resources and power, as we and biological beings have done for a long time.

To deal with incompleteness and uncertainty probabilistic Bayesian approaches have been pursued, with outstanding results. However, all these works, even if they propose probabilistic models, still rely on a classical computing paradigm that imposes a bottleneck on the performance and scalability. Improved and novel electronic devices have opened the spectrum of devices available for computation, such as GPUs, FPGAs, hybrid systems, allowing unconventional approaches to better explore parallelization and tackle power consumption. The flexibility of current reprogrammable logic devices provides a test bed for novel stochastic processors and unconventional computing.

The workshop will address recent advances and future directions of probabilistic computing for robotics, with keynote speakers on Bayesian inference for autonomous robots, and insights form computational biology, as well as presentations of submitted works, setting the floor for fruitful discussions and insights in this bridge topic.


Jorge Lobo, ISR - University of Coimbra (Contact person:

Joćo Filipe Ferreira, ISR - University of Coimbra

For further details please visit the website: