Contact

Emanuel Todorov
CSE 434
todorovcs.washington.edu
Artificial Intelligence, Machine Learning, Robotics
Areas of interest: 

Intelligent control in biology and engineering

Neuromuscular stochastic optimal control of a tendon-driven index finger model

E. Theodorou, E. Todorov, F. Valero-Cuevas, American Control Conference, 2011.

Aggregation methods for linearly-solvable MDPs

M. Zhong, E. Todorov, World Congress of the International Federation of Automatic Control, 2011.

Moving least-squares approximations for linearly-solvable stochastic optimal control problems

M. Zhong, E. Todorov, Journal of Control Theory Applications 9, 2011.

Inverse optimality design for biological movement systems

W. Li, E. Todorov, D. Liu, World Congress of the International Federation of Automatic Control, 2011.

A unifying framework for linearly-solvable control

K. Dvijotham, E. Todorov, Uncertainty in Artificial Intelligence, 2011.

Finding the most likely trajectories of optimally-controlled stochastic systems

E. Todorov, World Congress of the International Federation of Automatic Control, 2011.

High-order local dynamic programming

Y. Tassa, E. Todorov, IEEE Adaptive Dynamic Programming and Reinforcement Learning, 2011.

Inverse Optimal Control with Linearly Solvable MDPs

K. Dvijotham, E. Todorov, International Conference on Machine Learning, 2010.

Stochastic differential dynamic programming

E. Theodorou, Y. Tassa, E. Todorov, American Control Conference, 2010.

Position estimation and control of compact BLDC motors based on analog linear Hall effect sensors

A. Simpkins, E. Todorov, American Control Conference, 2010.

Identification and control of a pneumatic robot

E. Todorov, C. Hu, A. Simpkins, J. Movellan, IEEE BioRob, 2010.

A first optimal control solution for a complex, nonlinear, tendon driven neuromuscular finger model

E. Theodorou, E. Todorov, F. Valero-Cuevas, ASME Summer Bioengineering Conference, 2010.

Inverse optimal control with linearly-solvable MDPs

K. Dvijotham, T. E., International Conference on Machine Learning, 2010.

Policy gradients in linearly-solvable MDPs

E. Todorov, Advances in Neural Information Processing Systems, 2010.

Practical numerical methods for stochastic optimal control of biological systems in continuous time and space

A. Simpkins, E. Todorov, IEEE Adaptive Dynamic Programming and Reinforcement Learning, 2009.

Real-time motor control using recurrent neural networks

D. Huh, E. Todorov, IEEE Adaptive Dynamic Programming and Reinforcement Learning, 2009.

Structured variability of muscle activations supports the minimal intervention principle of motor control

F. Valero-Cuevas, M. Venkadesan, E. Todorov, Journal of Neurophysiology 102, 2009.

Iterative local dynamic programming

E. Todorov, Y. Tassa, IEEE Adaptive Dynamic Programming and Reinforcement Learning, 2009.

Efficient computation of optimal actions

Todorov, Proceedings of the National Academy of Science 106, 2009.

Hierarchical optimal control of a 7-DOF arm model

D. Liu, E. Todorov, IEEE Adaptive Dynamic Programming and Reinforcement Learning, 2009.

Compositionality of optimal control laws

E. Todorov, Advances in Neural Information Processing Systems, 2009.

Eigenfunction approximation methods for linearly-solvable optimal control problems

E. Todorov, IEEE Adaptive Dynamic Programming and Reinforcement Learning, 2009.

Optimal trade-off between exploration and exploitation

A. Simpkins, R. Callafon, E. Todorov, American Control Conference, 2008.

Parallels between sensory and motor information processing

E. Todorov, The Cognitive Neurosciences, 4th ed., MIT Press, 2008.

General duality between optimal control and estimation

E. Todorov, IEEE Conference on Decision and Control, 2008.

Probabilistic inference of multi-joint movements, skeletal parameters and marker attachments from diverse sensor data

E. Todorov, IEEE Transactions on Biomedical Engineering 54, 2007.

State estimation with finite signals-to-noise models via linear matrix inequalities

W. Li, R. Skelton, E. Todorov, Journal of Dynamic Systems, Measurement and Control 129, 2007.

Predicting reaching targets from human EEG

P. Hammon, S. Makeig, E. Poizner, V. Sa, IEEE Signal Processing Magazine 25, 2007.

Iterative linearization methods for approximately optimal control and estimation of non-linear stochastic systems

W. Li, E. Todorov, International Journal of Control 80, 2007.

Evidence for the flexible sensorimotor strategies predicted by optimal feedback control

D. Liu, E. Todorov, Journal of Neuroscience 27, 2007.

Optimal control theory

E. Todorov, Bayesian Brain: Probabilistic Approaches to Neural Coding, MIT Press, 2006.

Linearly-solvable Markov decision problems

E. Todorov, Advances in Neural Information Processing Systems, 2006.

Iterative optimal control and estimation design for nonlinear stochastic systems

W. Li, E. Todorov, IEEE Conference on Decision and Control, 2006.

Imitiation learning for reaching and grasping in virtual environments

N. Singh, E. Todorov, International Conference on Development and Learning, 2006.

Towards an integrated systems for estimating multi-joint movement from diverse sensor data

X. Pan, E. Todorov, W. Li, IEEE Engineering in Medicine and Biology, 2005.

Hierarchical feedback and learning for multi-joint arm movement control

W. Li, E. Todorov, X. Pan, IEEE Engineering in Medicine and Biology, 2005.

Stochastic optimal control and estimation methods adapted to the noise characteristics of the sensorimotor system

E. Todorov, Neural Computation 17, 2005.

From task parameters to motor synergies: A hierarchical framework for approximately-optimal control of redundant manipulators

E. Todorov, W. Li, X. Pan, Journal of Robotic Systems 22, 2005.

Estimation and control of systems with multiplicative noise via linear matrix inequalities

W. Li, R. Skelton, E. Todorov, American Control Conference, 2005.

A generalized iterative LQG method for locally-optimal feedback control of constrained nonlinear stochastic systems

E. Todorov, W. Li, American Control Conference, 2005.

Optimality principles in sensorimotor control

E. Todorov, Nature Neuroscience 7, 2004.

Iterative linear-quadratic regulator design for nonlinear biological movement systems

W. Li, E. Todorov, International Conference on Informatics in Control, Automation and Robotics, 2004.

Development of clinician-friendly software for musculoskeletal modeling and control

R. Davoodi, C. Urata, E. Todorov, G. Loeb, IEEE Engineering in Medicine and Biology, 2004.

Hierarchical optimal control of redundant biomechanical systems

W. Li, E. Todorov, X. Pan, IEEE Engineering in Medicine and Biology, 2004.

Analysis of the synergies underlying complex hand manipulation

E. Todorov, Z. Ghahramani, IEEE Engineering in Medicine and Biology, 2004.

Optimal control methods suitable for biomechanical systems

E. Todorov, W. Li, IEEE Engineering in Medicine and Biology, 2003.

Unsupervised learning of sensory-motor primitives

E. Todorov, Z. Ghahramani, IEEE Engineering in Medicine and Biology, 2003.

On the role of primary motor cortex in arm movement control

E. Todorov, Progress in Motor Control III, Human Kinetics, 2003.

A minimal intervention principle for coordinated movement

E. Todorov, M. Jordan, Advances in Neural Information Processing Systems, 2003.

Optimal feedback control as a theory of motor coordination

E. Todorov, M. Jordan, Nature Neuroscience 5, 2002.

Use of virtual environments in motor learning and rehabilitation

M. Holden, E. Todorov, Handbook of Virtual Environments, Lawrence Erlbaum Associates, 2002.

Cosine tuning minimizes motor errors

E. Todorov, Neural Computation 14, 2002.

A biomechanical model of the partially paralyzed human arm

R. Davoodi, I. Brown, E. Todorov, G. Loeb, Annual Conference of the International Functional Electric Stimulation Society, 2002.

One motor cortex, two different views

E. Todorov, Nature Neuroscience 3, 2000.

Direct cortical control of muscle activation in voluntary arm movements: A model

E. Todorov, Nature Neuroscience 3, 2000.

Virtual environment training improves motor performance in two patients with stroke

M. Holden, E. Todorov, J. Callahan, E. Bizzi, Neurology Report 23, 1999.

A local circuit approach to understanding integration of long-range inputs in primary visual cortex

D. Somers, E. Todorov, Cerebral Cortex 8, 1998.

Smoothness maximization along a predefined path accurately predicts the speed profiles of complex arm movements

E. Todorov, M. Jordan, Journal of Neurophysiology 80, 1998.

Modeling visual cortical contrast adaptation effects

E. Todorov, A. Siapas, D. Somers, S. Nelson, Computational Neuroscience, 1997.

Augmented feedback presented in a virtual environment accelerates learning of a difficult motor task

E. Todorov, R. Shadmehr, E. Bizzi, Journal of Motor Behavior 29, 1997.

A local circuit integration approach to understanding visual cortical receptive fields

D. Somers, E. Todorov, A. Siapas, Computational Neuroscience, 1997.

A model of recurrent interactions in primary visual cortex

E. Todorov, A. Siapas, D. Somers, Advances in Neural Information Processing Systems, 1997.

Variable gain control in local cortical circuitry supports context-dependent modulation by long-range connections

D. Somers, L. Toth, E. Todorov, Lateral Interactions in Cortex - Structure and Function, Online Book, 1996.

Factorial learning by clustering features

J. Tenenbaum, E. Todorov, Advances in Neural Information Processing Systems, 1995.

Catastrophic interference in human motor learning

T. Brashers-Krug, R. Shadmehr, E. Todorov, Advances in Neural Information Processing Systems, 1995.