no code implementations • 2 Dec 2023 • Cyrus Neary, Christian Ellis, Aryaman Singh Samyal, Craig Lennon, Ufuk Topcu
We propose and demonstrate a compositional framework for training and verifying reinforcement learning (RL) systems within a multifidelity sim-to-real pipeline, in order to deploy reliable and adaptable RL policies on physical hardware.
no code implementations • 2 Nov 2023 • Tichakorn Wongpiromsarn, Mahsa Ghasemi, Murat Cubuktepe, Georgios Bakirtzis, Steven Carr, Mustafa O. Karabag, Cyrus Neary, Parham Gohari, Ufuk Topcu
Formal methods refer to rigorous, mathematical approaches to system development and have played a key role in establishing the correctness of safety-critical systems.
no code implementations • 9 Sep 2023 • Cyrus Neary, Aryaman Singh Samyal, Christos Verginis, Murat Cubuktepe, Ufuk Topcu
We propose a framework for verifiable and compositional reinforcement learning (RL) in which a collection of RL subsystems, each of which learns to accomplish a separate subtask, are composed to achieve an overall task.
no code implementations • 10 Aug 2023 • Yunhao Yang, Cyrus Neary, Ufuk Topcu
We develop an algorithm that utilizes the knowledge from pretrained models to construct and verify controllers for sequential decision-making tasks, and to ground these controllers to task environments through visual observations.
no code implementations • 10 Jun 2023 • Franck Djeumou, Cyrus Neary, Ufuk Topcu
We present a framework and algorithms to learn controlled dynamics models using neural stochastic differential equations (SDEs) -- SDEs whose drift and diffusion terms are both parametrized by neural networks.
1 code implementation • 20 Jan 2023 • Bo Chen, Calvin Hawkins, Mustafa O. Karabag, Cyrus Neary, Matthew Hale, Ufuk Topcu
We synthesize policies that are robust to privacy by reducing the value of the total correlation.
no code implementations • 9 Jan 2023 • Adam J. Thorpe, Cyrus Neary, Franck Djeumou, Meeko M. K. Oishi, Ufuk Topcu
Our proposed approach incorporates prior knowledge of the system dynamics as a bias term in the kernel learning problem.
no code implementations • 4 Dec 2022 • Yunhao Yang, Jean-Raphaël Gaglione, Cyrus Neary, Ufuk Topcu
However, the textual outputs from GLMs cannot be formally verified or used for sequential decision-making.
1 code implementation • 1 Dec 2022 • Cyrus Neary, Ufuk Topcu
Toward the objective of learning composite models of such systems from data, we present i) a framework for compositional neural networks, ii) algorithms to train these models, iii) a method to compose the learned models, iv) theoretical results that bound the error of the resulting composite models, and v) a method to learn the composition itself, when it is not known a priori.
1 code implementation • 17 Jan 2022 • Mustafa O. Karabag, Cyrus Neary, Ufuk Topcu
In this work, we develop joint policies for cooperative multiagent systems that are robust to potential losses in communication.
1 code implementation • 14 Jan 2022 • Franck Djeumou, Cyrus Neary, Eric Goubault, Sylvie Putot, Ufuk Topcu
Neural ordinary differential equations (NODEs) -- parametrizations of differential equations using neural networks -- have shown tremendous promise in learning models of unknown continuous-time dynamical systems from data.
1 code implementation • 14 Sep 2021 • Franck Djeumou, Cyrus Neary, Eric Goubault, Sylvie Putot, Ufuk Topcu
The physics-informed constraints are enforced via the augmented Lagrangian method during the model's training.
1 code implementation • 7 Jun 2021 • Cyrus Neary, Christos Verginis, Murat Cubuktepe, Ufuk Topcu
We propose a framework for verifiable and compositional reinforcement learning (RL) in which a collection of RL subsystems, each of which learns to accomplish a separate subtask, are composed to achieve an overall task.
2 code implementations • 3 Jul 2020 • Cyrus Neary, Zhe Xu, Bo Wu, Ufuk Topcu
In cooperative multi-agent reinforcement learning, a collection of agents learns to interact in a shared environment to achieve a common goal.