Nicolas Heess
Informatics Forum, Room 2.51
10 Crichton Street
Edinburgh
EH8 9AB
n (dot) m (dot) o (dot) heess (at) sms (dot) ed (dot) ac (dot) uk
Research
I am a PhD student at the Institute for Adaptive and Neural
Computation, University of Edinburgh, UK, under the supervision of Prof. Chris
Williams. My research focusses on various aspects of machine
learning, in particular in the application of probabilistic models and
unsupervised learning to problems in computer vision and image
modeling. I am further interested in understanding information
processing in the brain. This includes questions related to neural
coding (especially in vision) and motion processing as well as the
application of machine learning techniques to neural data analysis.
Publications
Journal Articles
-
N. Le Roux*, N. Heess*, J. Shotton, J. Winn; Learning a generative
model of images by factoring appearance and shape;
Neural Computation 23(3): 593-650, 2011 (* joint first authorship)
-
N. Heess, W. Bair;
Direction Opponency, Not Quadrature, Is Key to
the 1/4 Cycle Preference for Apparent Motion in the Motion Energy
Model; Journal of Neuroscience, August 25, 2010,
30(34):11300-11304
Conference Papers
-
S.M. Eslami, N. Heess, J. Winn;
The Shape Boltzmann Machine: a Strong Model of Object Shape;
Computer Vision and Pattern Recognition (CVPR), Providence, USA, 2012
-
H.P. Saal, N. Heess, S. Vijayakumar;
Multimodal nonlinear filtering using Gauss-Hermite Quadrature;
European Conference on Machine Learning (ECML), Athens, Greece, 2011
-
N. Heess, N. Le Roux, J. Winn; Weakly Supervised Learning of Foreground-Background Segmentation using Masked RBMs; International Conference on Artificial Neural Networks (ICANN), Espoo, Finland, 2011
-
N. Heess, C.K.I. Williams, G.E. Hinton;
Learning generative texture models with extended Fields-of-experts
;
British Machine Vision Conference, London, UK, 2009 (oral
presentation; Supplementary
Material)
Conference Abstracts
-
N. Heess, W. Bair; Direction Opponency, not Quadrature, is Key to the
1/4 Cycle Preference for Apparent Motion in the Motion Energy Model;
Society for Neuroscience Annual Meeting, San Diego, USA, 2010 (oral
presentation)
-
N. Le Roux, N. Heess, J. Winn, J. Shotton;
Deep Segmentation Networks;
The Learning Workshop, Snowbird, USA, 2010 (oral presentation)
-
N. Heess, E. Blacker, D. McLelland, B. Ahmed, W. Bair; Spatial
integration in direction selective cortical neurons and the notion of
a fundamental spatial subunit; Society for Neuroscience Annual
Meeting, Washington, USA, 2008
-
N. Heess and C. Beckmann; Towards unconstrained fMRI: An investigation
of the variability induced by task or stimulus complexity;
Organization for Human Brain Mapping 12th Annual Meeting; Florence,
Italy, 2006
-
N. Heess and W. Bair; Modelling Adaptive Mechanisms for Motion
Processing in the Macaque Visual Cortex; Computational and Systems
Neuroscience (CoSyNe) Meeting; Abstract 62; Salt Lake City, USA, 2006
-
N. Heess and W. Bair; Motion gain control in direction selective cells
in macaque visual cortex; Society for Neuroscience Annual Meeting,
Abstract 136.7, Washington, USA, 2005 (oral presentation)
Patents
N. Le Roux, J. Winn, J. Shotton, N. Heess; Image Processing Using Masked Restricted Boltzmann Machines US patent application 12/535178; filed August 04, 2009 (patent application on IP.com)
Invited Talks
-
Sensory Coding and Stimulus Statistics; Lecture in the
computational neuroscience module of the MSc in Neuroscience
programme, University of Oxford; January 2008
-
Adaptive Motion Coding in Direction Selective
Neurons in Macaque Visual Cortex; The Rank Prize Funds, Mini Symposium
on Self-Calibrating Sensory Systems; Windermere, United Kingdom, 2007