2 code implementations • 7 Jul 2017 • Grady Williams, Paul Drews, Brian Goldfain, James M. Rehg, Evangelos A. Theodorou
We present an information theoretic approach to stochastic optimal control problems that can be used to derive general sampling based optimization schemes.
Robotics
no code implementations • ICML 2017 • Yunpeng Pan, Xinyan Yan, Evangelos A. Theodorou, Byron Boots
Sparse Spectrum Gaussian Processes (SSGPs) are a powerful tool for scaling Gaussian processes (GPs) to large datasets.
no code implementations • 16 Oct 2017 • Li Wang, Evangelos A. Theodorou, Magnus Egerstedt
The barrier certificates establish a non-conservative forward invariant safe region, in which high probability safety guarantees are provided based on the statistics of the Gaussian Process.
no code implementations • 27 Mar 2018 • Keuntaek Lee, Kamil Saigol, Evangelos A. Theodorou
We propose the use of Bayesian networks, which provide both a mean value and an uncertainty estimate as output, to enhance the safety of learned control policies under circumstances in which a test-time input differs significantly from the training set.
no code implementations • 12 Jun 2018 • Ching-An Cheng, Xinyan Yan, Evangelos A. Theodorou, Byron Boots
When the model oracle is learned online, these algorithms can provably accelerate the best known convergence rate up to an order.
no code implementations • 26 Apr 2019 • Keuntaek Lee, Gabriel Nakajima An, Viacheslav Zakharov, Evangelos A. Theodorou
The proposed information processing architecture is used to support a perceptual attention-based predictive control algorithm that leverages model predictive control (MPC), convolutional neural networks (CNNs), and uncertainty quantification methods.
no code implementations • 13 May 2019 • Grady Williams, Brian Goldfain, James M. Rehg, Evangelos A. Theodorou
We consider the problem of online adaptation of a neural network designed to represent vehicle dynamics.
no code implementations • 11 Jun 2019 • Marcus A. Pereira, Ziyi Wang, Tianrong Chen, Emily Reed, Evangelos A. Theodorou
We present a deep recurrent neural network architecture to solve a class of stochastic optimal control problems described by fully nonlinear Hamilton Jacobi Bellmanpartial differential equations.
no code implementations • 11 Jun 2019 • Ziyi Wang, Keuntaek Lee, Marcus A. Pereira, Ioannis Exarchos, Evangelos A. Theodorou
This paper presents a novel approach to numerically solve stochastic differential games for nonlinear systems.
1 code implementation • 28 Aug 2019 • Guan-Horng Liu, Evangelos A. Theodorou
In this article, we provide one possible way to align existing branches of deep learning theory through the lens of dynamical system and optimal control.
2 code implementations • 5 Oct 2019 • David D. Fan, Jennifer Nguyen, Rohan Thakker, Nikhilesh Alatur, Ali-akbar Agha-mohammadi, Evangelos A. Theodorou
We propose a framework which satisfies these constraints while allowing the use of deep neural networks for learning model uncertainties.
no code implementations • 7 Jan 2020 • Keuntaek Lee, Jason Gibson, Evangelos A. Theodorou
In this work, we couple a model predictive control (MPC) framework to a visual pipeline.
no code implementations • 5 Feb 2020 • David D. Fan, Ali-akbar Agha-mohammadi, Evangelos A. Theodorou
This uncertainty may come from errors in learning (due to a lack of data, for example), or may be inherent to the system.
no code implementations • ICLR 2021 • Guan-Horng Liu, Tianrong Chen, Evangelos A. Theodorou
Interpretation of Deep Neural Networks (DNNs) training as an optimal control problem with nonlinear dynamical systems has received considerable attention recently, yet the algorithmic development remains relatively limited.
no code implementations • 17 Apr 2020 • Keuntaek Lee, Bogdan Vlahov, Jason Gibson, James M. Rehg, Evangelos A. Theodorou
In this work, we present a method for obtaining an implicit objective function for vision-based navigation.
no code implementations • ICLR 2021 • Ioannis Exarchos, Marcus A. Pereira, Ziyi Wang, Evangelos A. Theodorou
In this work we propose the use of adaptive stochastic search as a building block for general, non-convex optimization operations within deep neural network architectures.
no code implementations • 17 Jul 2020 • Guan-Horng Liu, Tianrong Chen, Evangelos A. Theodorou
Connections between Deep Neural Networks (DNNs) training and optimal control theory has attracted considerable attention as a principled tool of algorithmic design.
no code implementations • 2 Sep 2020 • Marcus Aloysius Pereira, Ziyi Wang, Ioannis Exarchos, Evangelos A. Theodorou
This paper introduces a new formulation for stochastic optimal control and stochastic dynamic optimization that ensures safety with respect to state and control constraints.
no code implementations • 2 Sep 2020 • Ziyi Wang, Oswin So, Keuntaek Lee, Camilo A. Duarte, Evangelos A. Theodorou
We present a general framework for optimizing the Conditional Value-at-Risk for dynamical systems using stochastic search.
Distributional Reinforcement Learning Optimization and Control Robotics
no code implementations • 28 Sep 2020 • Lin Song, Neng Wan, Aditya Gahlawat, Naira Hovakimyan, Evangelos A. Theodorou
The proposed approach achieves both the compositionality and optimality of control actions simultaneously within the cooperative MAS framework in both discrete- and continuous-time in a sample-efficient manner, which reduces the burden of re-computation of the optimal control solutions for the new task on the MASs.
no code implementations • 30 Sep 2020 • Neng Wan, Aditya Gahlawat, Naira Hovakimyan, Evangelos A. Theodorou, Petros G. Voulgaris
Local control actions that rely only on agents' local observations are designed to optimize the joint cost functions of subsystems.
no code implementations • 21 Nov 2020 • Tianrong Chen, Ziyi Wang, Ioannis Exarchos, Evangelos A. Theodorou
We showcase superior performance of our framework over the state-of-the-art deep fictitious play algorithm on an inter-bank lending/borrowing problem in terms of multiple metrics.
no code implementations • 18 Feb 2021 • Neng Wan, Aditya Gahlawat, Naira Hovakimyan, Evangelos A. Theodorou, Petros G. Voulgaris
Distributed algorithms for both discrete-time and continuous-time linearly solvable optimal control (LSOC) problems of networked multi-agent systems (MASs) are investigated in this paper.
no code implementations • 20 Feb 2021 • Hassan Almubarak, Nader Sadegh, Evangelos A. Theodorou
The development enforces safety by means of barrier functions used in optimization through the construction of barrier states (BaS) which are embedded in the control system's model.
no code implementations • 1 Apr 2021 • Ziyi Wang, Oswin So, Jason Gibson, Bogdan Vlahov, Manan S. Gandhi, Guan-Horng Liu, Evangelos A. Theodorou
In this paper, we provide a generalized framework for Variational Inference-Stochastic Optimal Control by using thenon-extensive Tsallis divergence.
no code implementations • 8 May 2021 • Guan-Horng Liu, Tianrong Chen, Evangelos A. Theodorou
The connection between training deep neural networks (DNNs) and optimal control theory (OCT) has attracted considerable attention as a principled tool of algorithmic design.
no code implementations • 29 Jun 2021 • Hassan Almubarak, Evangelos A. Theodorou, Nader Sadegh
This work proposes an optimal safe controller minimizing an infinite horizon cost functional subject to control barrier functions (CBFs) safety conditions.
no code implementations • 25 Jul 2021 • David D. Fan, Sharmita Dey, Ali-akbar Agha-mohammadi, Evangelos A. Theodorou
One of the main challenges in autonomous robotic exploration and navigation in unknown and unstructured environments is determining where the robot can or cannot safely move.
no code implementations • 1 Sep 2021 • Marcus A. Pereira, Camilo A. Duarte, Ioannis Exarchos, Evangelos A. Theodorou
In this paper, we introduce a novel deep learning based solution to the Powered-Descent Guidance (PDG) problem, grounded in principles of nonlinear Stochastic Optimal Control (SOC) and Feynman-Kac theory.
no code implementations • 21 Sep 2021 • Lin Song, Neng Wan, Aditya Gahlawat, Chuyuan Tao, Naira Hovakimyan, Evangelos A. Theodorou
The control action composition is achieved by taking a weighted mixture of the existing controllers according to the contribution of each component task.
1 code implementation • NeurIPS 2021 • Guan-Horng Liu, Tianrong Chen, Evangelos A. Theodorou
We propose a novel second-order optimization framework for training the emerging deep continuous-time models, specifically the Neural Ordinary Differential Equations (Neural ODEs).
1 code implementation • ICLR 2022 • Tianrong Chen, Guan-Horng Liu, Evangelos A. Theodorou
However, it remains unclear whether the optimization principle of SB relates to the modern training of deep generative models, which often rely on constructing log-likelihood objectives. This raises questions on the suitability of SB models as a principled alternative for generative applications.
Ranked #44 on Image Generation on CIFAR-10
no code implementations • 4 Nov 2021 • Hassan Almubarak, Evangelos A. Theodorou, Nader Sadegh
The proposed method leverages a game theoretic differential dynamic programming approach with barrier states to handle parametric and non-parametric uncertainties in safety-critical control systems.
no code implementations • 17 Jan 2022 • Keuntaek Lee, David Isele, Evangelos A. Theodorou, Sangjae Bae
It can be difficult to autonomously produce driver behavior so that it appears natural to other traffic participants.
no code implementations • 22 Feb 2022 • Marcus A. Pereira, Augustinos D. Saravanos, Oswin So, Evangelos A. Theodorou
In this work, we propose a novel safe and scalable decentralized solution for multi-agent control in the presence of stochastic disturbances.
no code implementations • 5 Apr 2022 • Tianrong Chen, Ziyi Wang, Evangelos A. Theodorou
Our approach relies on the probabilistic representation of the solution of the Hamilton-Jacobi-Bellman partial differential equation.
no code implementations • 27 Jul 2022 • Augustinos D. Saravanos, Yuichiro Aoyama, Hongchang Zhu, Evangelos A. Theodorou
The aim of this work is to suggest architectures that inherit the computational efficiency and scalability of Differential Dynamic Programming (DDP) and the distributed nature of the Alternating Direction Method of Multipliers (ADMM).
1 code implementation • 20 Sep 2022 • Guan-Horng Liu, Tianrong Chen, Oswin So, Evangelos A. Theodorou
In this work, we aim at solving a challenging class of MFGs in which the differentiability of these interacting preferences may not be available to the solver, and the population is urged to converge exactly to some desired distribution.
no code implementations • 30 Sep 2022 • Oswin So, Gongjie Li, Evangelos A. Theodorou, Molei Tao
Incorporating the Hamiltonian structure of physical dynamics into deep learning models provides a powerful way to improve the interpretability and prediction accuracy.
no code implementations • 1 Dec 2022 • Hassan Almubarak, Manan Gandhi, Yuichiro Aoyama, Nader Sadegh, Evangelos A. Theodorou
We derive the barrier state dynamics utilizing the GP posterior, which is used to construct a safety embedded Gaussian process dynamical model (GPDM).
1 code implementation • 12 Feb 2023 • Guan-Horng Liu, Arash Vahdat, De-An Huang, Evangelos A. Theodorou, Weili Nie, Anima Anandkumar
We propose Image-to-Image Schr\"odinger Bridge (I$^2$SB), a new class of conditional diffusion models that directly learn the nonlinear diffusion processes between two given distributions.
1 code implementation • NeurIPS 2023 • Guan-Horng Liu, Tianrong Chen, Evangelos A. Theodorou, Molei Tao
In this work, we propose Mirror Diffusion Models (MDM), a new class of diffusion models that generate data on convex constrained sets without losing any tractability.
1 code implementation • 3 Oct 2023 • Guan-Horng Liu, Yaron Lipman, Maximilian Nickel, Brian Karrer, Evangelos A. Theodorou, Ricky T. Q. Chen
Modern distribution matching algorithms for training diffusion or flow models directly prescribe the time evolution of the marginal distributions between two boundary distributions.
no code implementations • 10 Oct 2023 • Hassan Almubarak, Nader Sadegh, Evangelos A. Theodorou
The proposition is that the control problem is now transformed to designing a control law for the new, unconstrained, system in which the barrier state is driven to stay bounded while achieving other performance objectives.
no code implementations • 11 Oct 2023 • Tianrong Chen, Jiatao Gu, Laurent Dinh, Evangelos A. Theodorou, Josh Susskind, Shuangfei Zhai
In this work, we introduce a novel generative modeling framework grounded in \textbf{phase space dynamics}, where a phase space is defined as {an augmented space encompassing both position and velocity.}
no code implementations • 12 Nov 2023 • Valentin De Bortoli, Guan-Horng Liu, Tianrong Chen, Evangelos A. Theodorou, Weilie Nie
In this paper, we highlight that while flow and bridge matching processes preserve the information of the marginal distributions, they do \emph{not} necessarily preserve the coupling information unless additional, stronger optimality conditions are met.
no code implementations • 9 Apr 2024 • Yuchen Zhu, Tianrong Chen, Evangelos A. Theodorou, Xie Chen, Molei Tao
This article considers the generative modeling of the states of quantum systems, and an approach based on denoising diffusion model is proposed.
no code implementations • 20 Apr 2024 • Chenru Duan, Guan-Horng Liu, Yuanqi Du, Tianrong Chen, Qiyuan Zhao, Haojun Jia, Carla P. Gomes, Evangelos A. Theodorou, Heather J. Kulik
The RMSD and barrier height error is further improved by roughly 25% through pretraining React-OT on a large reaction dataset obtained with a lower level of theory, GFN2-xTB.