Rajesh P. N. Rao
Associate Professor, CSE, University of Washington
Faculty Member, Neurobiology & Behavior Program, University of Washington
Adjunct Associate Professor, EE, University of Washington
Faculty Member, Faculty of 1000 Biology, Theoretical Neuroscience Section
Sloan Postdoctoral Fellow, Salk Institute (1997-2000)
Ph.D., University of Rochester, 1998
Research Interests
Computational Neuroscience and Brain-Computer Interfaces
The primary goal of my research is to discover the computational
principles underlying the brain's remarkable ability to learn, process
and store information, and to apply this knowledge to the task of
building adaptive robotic systems and brain-computer interfaces
(BCIs). How does the brain learn efficient representations of objects
and events occurring in the natural environment? What are the
algorithms that allow useful sensorimotor behaviors to be learned?
What computational mechanisms allow the brain to adapt to changing
circumstances and remain fault-tolerant and robust? How can the
knowledge gained through computational studies of the brain be used in
biomedical applications such as BCIs for the disabled? My students and
I are investigating these questions using a combination of
probabilistic techniques, computer simulations, and collaborative
neurobiological experiments. Such an interdisciplinary approach has
provided functional interpretations of several otherwise puzzling
neurobiological properties while at the same time suggesting
biologically-inspired solutions to problems in computer vision,
robotics and artificial intelligence.
Humanoid Robots that Learn from Humans
We are developing new
methods that allow a humanoid robot to learn new actions and skills
from a human teacher in much the same way that human infants and adults
learn through observation and experimentation. Such an approach
opens the door to a potentially powerful way of programming
general-purpose humanoid robots--through human demonstration, obviating the need for complex physics-based models and explicit programming of behaviors.
Analysis of the 4500-year-old Indus Script
Despite a large
number of attempts, the script of the Indus civilization (circa
2500-1900 BC) remains undeciphered. The absence of a multilingual
"Rosetta stone" as well as our lack of knowledge of the underlying
language have stymied decipherment efforts. Rather than attempting to
ascribe meaning to the inscriptions, we are applying statistical
techniques from the fields of machine learning, information theory,
and computational linguistics to first gain an understanding of the
sequential structure of the script. The goal is to discover the
grammatical rules that govern the sequencing of signs in the script,
with the hope that such rules will aid future
decipherment efforts.
Press coverage
Indus Script:
Brain-Controlled Humanoid Robot:
Awards
Publications
Books
Selected Articles
Analysis of the 4500-year-old Indus Script:
Computational Neuroscience and Brain-Computer Interfaces:
-
Control of a humanoid robot by a noninvasive brain-computer interface in humans (Featured cover article, with C. Bell et al., J Neural Eng, 5(2):214-20, 2008). One of the seven Highlights of 2008. See also Press Coverage above.
- Online electromyographic control of a robotic prosthesis (with P. Shenoy et al., IEEE Trans Biomed Eng, 55(3):1128-35, 2008)
- Generalized features for electrocorticographic BCIs (with P. Shenoy et al., IEEE Trans Biomed Eng., 55(1):273-80, 2008)
- Beyond the gamma band: the role of high-frequency features in movement classification (with K. Miller et al., IEEE Trans Biomed Eng, 55(5):1634-7, 2008)
- Feasibility and pragmatics of classifying working memory load with an electroencephalograph (Best Paper Honorable Mention, with D. Grimes et al., Proceedings of the 2008 Conference on Computer-Human Interaction (CHI), 835-844, 2008)
- Real-time functional brain mapping using electrocorticography (with K. Miller et al., Neuroimage, 37(2):504-507, 2007).
- Spectral changes in cortical surface potentials during motor
movement (with K. Miller et al., The Journal of Neuroscience, 27(9):2424-32, 2007)
- Electrocorticography-based brain computer interface -- the Seattle experience (with Leuthardt et al., IEEE Trans Neural Syst Rehab Eng 14(2):194-8, 2006)
- Towards adaptive classification for BCI (with Shenoy et al., J Neural Eng 3(1):R13-23, 2006)
- Neural models of Bayesian belief propagation (Bayesian Brain, 2007)
- Bayesian inference and attention in the visual cortex (Neuroreport 16(16), 1843-1848, 2005)
- Hierarchical Bayesian inference in networks of spiking neurons (Advances in NIPS 17, 2005)
- Dynamic Bayesian networks for brain-computer interfaces (with P. Shenoy, Advances in NIPS 17, 2005)
- Bayesian computation in recurrent neural circuits (Neural Computation, 16(1), 1-38, 2004)
- A Bayesian model of imitation in infants and robots (with A. Shon and A. Meltzoff, Imitation and Social Learning in Robots, Humans, and Animals, Cambridge University Press, 2005)
- Probabilistic models of attention based on iconic representations and predictive coding (with Dana Ballard, Neurobiology of Attention, 2004)
- Self-organizing neural systems based on predictive learning (with Terry Sejnowski, Phil. Trans. Royal Soc. Lond. A, Vol. 361, 2003)
- Imitation learning in infants and robots (with Andy Meltzoff, Proc. of AISB'03, 2003)
- A Bilinear Model for Sparse Coding (with D. Grimes, Advances in NIPS 15, 2003)
- Learning Temporal Patterns by Redistribution of Synaptic Efficacy (with A. Shon, Neurocomputing, 2003)
- Receptive Field (invited review article) (Encyclopedia of the Human Brain, 2002)
- Models of Attention (invited review article) (Encyclopedia of Cognitive Science, 2002)
- Eye Movements in Iconic Visual Search (Vision Research 42(11), 1447-1463, 2002)
- Awakening a sleeping cat: Invited review of Information Theory and the Brain (Neural Networks, 2002)
- Spike Timing Dependent Hebbian Plasticity as Temporal Difference Learning (Neural Computation, 13(10), 2221-2237, 2001)
- News and Views by Peter Dayan (Trends in Cog. Sci., 6(3), 105-106, 2002)
- Optimal Smoothing in Visual Motion Perception (Neural Computation, 13(6), 1243-1253, 2001)
- Predictive Learning of Temporal Sequences in Neocortical Circuits (Complexity in Bio. Info. Processing, 2001)
- Reliability of Spike Timing in Cortical Neurons (J. Neurophysiology, 85(4), 1782-1787, 2001)
- Neural Circuits in Silicon [News & Views] (Nature, 405, 891-892, 2000)
- Predictive Sequence Learning in Recurrent Neocortical Circuits (Advances in NIPS 12, 164-170, 2000)
- Learning to Maximize Rewards: Review of the book "Reinforcement Learning" (Neural Networks, 13(1), pp. 135-137, 2000)
- Predictive Coding in the Visual Cortex (Nature Neuroscience, 2(1), 79-87, 1999)
- An Optimal Estimation Approach to Visual Perception and Learning (Vision Research, 39(11), 1963-1989, 1999)
- Learning Lie Groups for Invariant Visual Perception (Advances in NIPS 11, pp. 810-816, 1999)
-
What/Where Networks & Local Receptive Fields for Transformation Estimation (Network, 9(2), pp. 219-234, 1998)
-
Visual Attention during Recognition (Advances in NIPS 10, pp. 80-86, 1998)
- Kalman
Filter Model of the Visual Cortex (Neural Computation, 9(4), pp. 721-763, 1997)
- Deictic Codes for the Embodiment of Cognition (Behav. and Brain Sciences, 20(4), pp. 723-767, 1997)
Robotics, Machine Learning, and Vision:
- Learning nonparametric policies by imitation (with D. Grimes, Proceedings of the 2008 IEEE International Conference on Intelligent Robots and Systems (IROS), 2008)
- Learning full-body motions from monocular vision: Dynamic imitation in a humanoid robot (with J. Cole and D. Grimes, Proceedings of the 2007 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 240-246, 2007)
- Learning to walk through imitation (with R. Chalodhorn et al., Proceedings of the International Joint Conference on Artificial Intelligence (IJCAI'07), 2084-2090, 2007)
- Imitation learning using graphical models (with D. Verma, Proceedings of the 2007 European Conference on Machine Learning (ECML), 757-764, 2007)
- Towards a real-time Bayesian imitation system for a humanoid robot (with A. Shon and J. Storz, Proceedings of the 2007 International Conference on Robotics and Automation (ICRA), 2847-2852, 2007)
- Active imitation learning (with A. Shon and D. Verma, Proceedings of the 2007 Conference of the American Association for Artificial Intelligence (AAAI), 756-762, 2007)
- Learning nonparametric models for probabilistic imitation (with D. Grimes and D. Rashid, Advances in Neural Information Processing Systems 19 (NIPS'06), 521-528, 2007)
- Learning the Lie groups of visual invariance (with X. Miao, Neural Computation, 19(10):2665-2693, 2007)
- Planning and acting in uncertain environments using probabilistic inference (with D. Verma, Proceedings of IEEE/RSJ Int. Conf. on Intelligent Robots and Systems (IROS), 2006)
- Dynamic imitation in a humanoid robot through nonparametric probabilistic inference (with Grimes and Chalodhorn, Proceedings of the 2006 Robotics Science and Systems Conference, 2006)
- A probabilistic model of gaze imitation and shared attention (with Hoffman et al., Neural Networks 19(3):299-310, 2006)
- Learning Dynamic Humanoid Motion using Predictive Control in Low Dimensional Subspaces, Chalodhorn, R., Grimes, D. B., Maganis, G. and Rao, R. P. N., Proceeding of the IEEE-RAS/RSJ International Conference on Humanoid Robots, Dec 2005 (Humanoids 2005)
- Learning Humanoid Motion Dynamics through Sensory-Motor Mapping in Reduced Dimensional Spaces, Chalodhorn, R., Grimes, D. B., Maganis, G., Rao, R. P. N., and Asada, M., Proceeding of the IEEE International Conference on Robotics and Automation, May 2006 (ICRA 2006)
- Goal-based imitation as probabilistic inference over graphical models (Advances in NIPS 18, 2006)
- Learning shared latent structure for image synthesis and robotic imitation (Advances in NIPS 18, 2006)
- Bilinear sparse coding for invariant vision (with D. Grimes, Neural Computation, 17(1), 47-73, 2005)
-
Graphical models for planning and imitation (with D. Verma, Technical Report 2005-02-01, February 2005)
- A probabilistic framework for model-based imitation learning (with A. Shon, D. Grimes, and C. Baker, Proceedings of the 26th Annual Meeting of the Cognitive Science Society, 2004)
- Probabilistic bilinear models for appearance-based vision (with David Grimes and Aaron Shon, Proc. of ICCV'03, 2003)
Prior Work:
- Learning in Mobile Robots:
-
Learning Spatiotemporal Receptive Fields from Natural Images (Tech Report 97.4,Dept of Comp Sci, Univ of Rochester, 1997)
-
Dynamic Appearance-Based Recognition (Proceedings of CVPR'97, pp. 540-546, 1997)
- Eye
Movements in Visual Cognition (Technical Report, 1997)
- Robust Kalman Filters (Technical Report, 1997)
- Kalman Filter Models for Invariant Recognition, Motion, and Stereo (Technical Report, 1996)
- Object-Centered Neglect (Computational Neuroscience: Trends in Research 1997)
- Modeling Human Eye Movements in Visual Search (Advances in NIPS 8, pp. 830-836, 1996)
- Face Recognition using Natural Basis Functions (Proceedings of IJCAI*95, pp. 10-17, 1995)
-
An Active Vision Architecture based on Iconic Representations (Artificial Intelligence 78, pp. 461-505, 1995)
-
Learning Saccadic Eye Movements using Multiscale Spatial Filters (Advances in NIPS 7, pp. 893-900, 1995)
Other Activities:
- Band member, The Smashing Notkins (2002-2003). Official Band Photo (copyright: Melissa Garcia, 2003)
Rajesh Rao
Computer Science & Engineering
566 Allen Center
University of Washington
Box 352350
Seattle, WA 98195-2350
Phone: 206-685-9141
Fax: 206-543-2969
WWW: http://www.cs.washington.edu/homes/rao
e-mail: rao[at]cs[dot]washington[dot]edu
