There has been a recent upsurge in interest in Gaussian processes for solving a variety of machine learning problems. Simultaneously there have been recent rapid developments in efficient approximation techniques for Gaussian processes and novel unifying theories of existing approximation techniques. This workshop will bring experts in Gaussian processes together with individuals who are using Gaussian processes at the forefront of research in their fields. We will start with a set of talks that will relate the latest developments in Gaussian processes (speakers include Chris Williams, Carl Rasmussen, David MacKay and Ed Snelson) while the second portion of talks will discuss application fields where Gaussian processes are being applied as state-of-the-art technologies: Robotics, Graphics and Vision (speakers include Brian Ferris, Aaron Hertzmann and Raquel Urtasun).