Hierarchical Decomposition of Nonlinear Dynamics and Control for System Identification and Policy Distillation

• URL: http://arxiv.org/abs/2005.01432v2
• Date: Tue, 12 May 2020 14:54:33 GMT
• Title: Hierarchical Decomposition of Nonlinear Dynamics and Control for System Identification and Policy Distillation
• Authors: Hany Abdulsamad and Jan Peters
• Abstract summary: Current trends in reinforcement learning (RL) focus on complex representations of dynamics and policies. We take inspiration from the control community and apply the principles of hybrid switching systems in order to break down complex dynamics into simpler components.
• Score: 39.83837705993256
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: The control of nonlinear dynamical systems remains a major challenge for autonomous agents. Current trends in reinforcement learning (RL) focus on complex representations of dynamics and policies, which have yielded impressive results in solving a variety of hard control tasks. However, this new sophistication and extremely over-parameterized models have come with the cost of an overall reduction in our ability to interpret the resulting policies. In this paper, we take inspiration from the control community and apply the principles of hybrid switching systems in order to break down complex dynamics into simpler components. We exploit the rich representational power of probabilistic graphical models and derive an expectation-maximization (EM) algorithm for learning a sequence model to capture the temporal structure of the data and automatically decompose nonlinear dynamics into stochastic switching linear dynamical systems. Moreover, we show how this framework of switching models enables extracting hierarchies of Markovian and auto-regressive locally linear controllers from nonlinear experts in an imitation learning scenario.

Related papers

• Discovering Governing equations from Graph-Structured Data by Sparse Identification of Nonlinear Dynamical Systems [0.27624021966289597]
  We develop a new method called Sparse Identification of Dynamical Systems from Graph-structured data (SINDyG) SINDyG incorporates the network structure into sparse regression to identify model parameters that explain the underlying network dynamics.
  arXiv  Detail & Related papers  (2024-09-02T17:51:37Z)
• Learning System Dynamics without Forgetting [60.08612207170659]
  Predicting trajectories of systems with unknown dynamics is crucial in various research fields, including physics and biology. We present a novel framework of Mode-switching Graph ODE (MS-GODE), which can continually learn varying dynamics. We construct a novel benchmark of biological dynamic systems, featuring diverse systems with disparate dynamics.
  arXiv  Detail & Related papers  (2024-06-30T14:55:18Z)
• Data driven modeling for self-similar dynamics [1.0790314700764785]
  We introduce a multiscale neural network framework that incorporates self-similarity as prior knowledge. For deterministic dynamics, our framework can discern whether the dynamics are self-similar. Our method can identify the power law exponents in self-similar systems.
  arXiv  Detail & Related papers  (2023-10-12T12:39:08Z)
• Capturing Actionable Dynamics with Structured Latent Ordinary Differential Equations [68.62843292346813]
  We propose a structured latent ODE model that captures system input variations within its latent representation. Building on a static variable specification, our model learns factors of variation for each input to the system, thus separating the effects of the system inputs in the latent space.
  arXiv  Detail & Related papers  (2022-02-25T20:00:56Z)
• Model-Based Reinforcement Learning via Stochastic Hybrid Models [39.83837705993256]
  This paper adopts a hybrid-system view of nonlinear modeling and control. We consider a sequence modeling paradigm that captures the temporal structure of the data. We show that these time-series models naturally admit a closed-loop extension that we use to extract local feedback controllers.
  arXiv  Detail & Related papers  (2021-11-11T14:05:46Z)
• Reduced Order Dynamical Models For Complex Dynamics in Manufacturing and Natural Systems Using Machine Learning [0.0]
  This work develops reduced-order models of manufacturing and natural systems using a machine learning (ML) approach. The approach is demonstrated on an entire soybean-oil to soybean-diesel process plant and a lake system. Results show that the method identifies a high accuracy linear ODE models for the process plant, reflective of underlying linear stoichiometric mechanisms and mass balance driving the dynamics.
  arXiv  Detail & Related papers  (2021-10-15T18:44:27Z)
• Supervised DKRC with Images for Offline System Identification [77.34726150561087]
  Modern dynamical systems are becoming increasingly non-linear and complex. There is a need for a framework to model these systems in a compact and comprehensive representation for prediction and control. Our approach learns these basis functions using a supervised learning approach.
  arXiv  Detail & Related papers  (2021-09-06T04:39:06Z)
• A purely data-driven framework for prediction, optimization, and control of networked processes: application to networked SIS epidemic model [0.8287206589886881]
  We develop a data-driven framework based on operator-theoretic techniques to identify and control nonlinear dynamics over large-scale networks. The proposed approach requires no prior knowledge of the network structure and identifies the underlying dynamics solely using a collection of two-step snapshots of the states.
  arXiv  Detail & Related papers  (2021-08-01T03:57:10Z)
• Controlling nonlinear dynamical systems into arbitrary states using machine learning [77.34726150561087]
  We propose a novel and fully data driven control scheme which relies on machine learning (ML) Exploiting recently developed ML-based prediction capabilities of complex systems, we demonstrate that nonlinear systems can be forced to stay in arbitrary dynamical target states coming from any initial state. Having this highly flexible control scheme with little demands on the amount of required data on hand, we briefly discuss possible applications that range from engineering to medicine.
  arXiv  Detail & Related papers  (2021-02-23T16:58:26Z)
• Learning Stable Deep Dynamics Models [91.90131512825504]
  We propose an approach for learning dynamical systems that are guaranteed to be stable over the entire state space. We show that such learning systems are able to model simple dynamical systems and can be combined with additional deep generative models to learn complex dynamics.
  arXiv  Detail & Related papers  (2020-01-17T00:04:45Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.